| { | |
| "best_metric": 1.9983205795288086, | |
| "best_model_checkpoint": "/home/datta0/models/lora_final/Mistral-7B-v0.3_pct_ortho_r16/checkpoint-16", | |
| "epoch": 0.9983071342200726, | |
| "eval_steps": 8, | |
| "global_step": 387, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0025796049979846837, | |
| "grad_norm": 7.310375690460205, | |
| "learning_rate": 1.25e-05, | |
| "loss": 2.1533, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010318419991938735, | |
| "grad_norm": 7.721648216247559, | |
| "learning_rate": 5e-05, | |
| "loss": 2.0899, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.02063683998387747, | |
| "grad_norm": 4.991214275360107, | |
| "learning_rate": 0.0001, | |
| "loss": 1.9566, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.02063683998387747, | |
| "eval_loss": 2.0118165016174316, | |
| "eval_runtime": 141.9765, | |
| "eval_samples_per_second": 1.726, | |
| "eval_steps_per_second": 0.866, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.030955259975816204, | |
| "grad_norm": 4.564762592315674, | |
| "learning_rate": 9.997251843068762e-05, | |
| "loss": 2.0017, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04127367996775494, | |
| "grad_norm": 4.318169593811035, | |
| "learning_rate": 9.989010393221656e-05, | |
| "loss": 2.0191, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04127367996775494, | |
| "eval_loss": 1.9983205795288086, | |
| "eval_runtime": 142.0598, | |
| "eval_samples_per_second": 1.725, | |
| "eval_steps_per_second": 0.866, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.051592099959693674, | |
| "grad_norm": 4.341538429260254, | |
| "learning_rate": 9.97528470997769e-05, | |
| "loss": 2.1128, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06191051995163241, | |
| "grad_norm": 3.7456655502319336, | |
| "learning_rate": 9.956089881469482e-05, | |
| "loss": 2.0779, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06191051995163241, | |
| "eval_loss": 2.0212316513061523, | |
| "eval_runtime": 139.6844, | |
| "eval_samples_per_second": 1.754, | |
| "eval_steps_per_second": 0.881, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07222893994357114, | |
| "grad_norm": 3.6751935482025146, | |
| "learning_rate": 9.931447007857432e-05, | |
| "loss": 2.0376, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.08254735993550988, | |
| "grad_norm": 3.707818031311035, | |
| "learning_rate": 9.901383178135113e-05, | |
| "loss": 2.0339, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08254735993550988, | |
| "eval_loss": 2.0204849243164062, | |
| "eval_runtime": 266.838, | |
| "eval_samples_per_second": 0.918, | |
| "eval_steps_per_second": 0.461, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.09286577992744861, | |
| "grad_norm": 3.904897689819336, | |
| "learning_rate": 9.865931440351337e-05, | |
| "loss": 2.0485, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10318419991938735, | |
| "grad_norm": 3.803988456726074, | |
| "learning_rate": 9.825130765281668e-05, | |
| "loss": 2.0429, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10318419991938735, | |
| "eval_loss": 2.0132498741149902, | |
| "eval_runtime": 135.71, | |
| "eval_samples_per_second": 1.805, | |
| "eval_steps_per_second": 0.906, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11350261991132608, | |
| "grad_norm": 4.293553829193115, | |
| "learning_rate": 9.779026003589304e-05, | |
| "loss": 2.0392, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.12382103990326482, | |
| "grad_norm": 4.716602802276611, | |
| "learning_rate": 9.727667836522407e-05, | |
| "loss": 2.0601, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12382103990326482, | |
| "eval_loss": 2.0219218730926514, | |
| "eval_runtime": 138.9003, | |
| "eval_samples_per_second": 1.764, | |
| "eval_steps_per_second": 0.886, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.13413945989520354, | |
| "grad_norm": 4.0252580642700195, | |
| "learning_rate": 9.6711127202021e-05, | |
| "loss": 2.0285, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.1444578798871423, | |
| "grad_norm": 3.81237530708313, | |
| "learning_rate": 9.609422823562345e-05, | |
| "loss": 2.041, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.1444578798871423, | |
| "eval_loss": 2.017057180404663, | |
| "eval_runtime": 136.05, | |
| "eval_samples_per_second": 1.801, | |
| "eval_steps_per_second": 0.904, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.154776299879081, | |
| "grad_norm": 4.155681610107422, | |
| "learning_rate": 9.542665960009959e-05, | |
| "loss": 2.0558, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.16509471987101976, | |
| "grad_norm": 3.761030435562134, | |
| "learning_rate": 9.470915512879852e-05, | |
| "loss": 2.0602, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.16509471987101976, | |
| "eval_loss": 2.0229883193969727, | |
| "eval_runtime": 136.2202, | |
| "eval_samples_per_second": 1.799, | |
| "eval_steps_per_second": 0.903, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.17541313986295848, | |
| "grad_norm": 4.972229480743408, | |
| "learning_rate": 9.394250354767467e-05, | |
| "loss": 2.0754, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.18573155985489723, | |
| "grad_norm": 3.9014289379119873, | |
| "learning_rate": 9.312754760827061e-05, | |
| "loss": 2.0341, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.18573155985489723, | |
| "eval_loss": 2.0311481952667236, | |
| "eval_runtime": 255.6823, | |
| "eval_samples_per_second": 0.958, | |
| "eval_steps_per_second": 0.481, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.19604997984683595, | |
| "grad_norm": 4.544624328613281, | |
| "learning_rate": 9.226518316131176e-05, | |
| "loss": 2.0667, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.2063683998387747, | |
| "grad_norm": 4.004471778869629, | |
| "learning_rate": 9.1356358171931e-05, | |
| "loss": 2.0378, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2063683998387747, | |
| "eval_loss": 2.0319135189056396, | |
| "eval_runtime": 138.1562, | |
| "eval_samples_per_second": 1.773, | |
| "eval_steps_per_second": 0.89, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.21668681983071342, | |
| "grad_norm": 4.80974006652832, | |
| "learning_rate": 9.040207167760586e-05, | |
| "loss": 2.0514, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.22700523982265217, | |
| "grad_norm": 4.184769153594971, | |
| "learning_rate": 8.940337268995385e-05, | |
| "loss": 2.0961, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.22700523982265217, | |
| "eval_loss": 2.0402026176452637, | |
| "eval_runtime": 141.956, | |
| "eval_samples_per_second": 1.726, | |
| "eval_steps_per_second": 0.866, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.2373236598145909, | |
| "grad_norm": 4.9076409339904785, | |
| "learning_rate": 8.836135904159302e-05, | |
| "loss": 2.0838, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.24764207980652964, | |
| "grad_norm": 5.397791385650635, | |
| "learning_rate": 8.727717617933544e-05, | |
| "loss": 2.106, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.24764207980652964, | |
| "eval_loss": 2.020820140838623, | |
| "eval_runtime": 136.0024, | |
| "eval_samples_per_second": 1.801, | |
| "eval_steps_per_second": 0.904, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.25796049979846836, | |
| "grad_norm": 4.622895240783691, | |
| "learning_rate": 8.615201590504017e-05, | |
| "loss": 2.0765, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2682789197904071, | |
| "grad_norm": 4.172301769256592, | |
| "learning_rate": 8.498711506550983e-05, | |
| "loss": 2.1219, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.2682789197904071, | |
| "eval_loss": 2.0327656269073486, | |
| "eval_runtime": 126.7113, | |
| "eval_samples_per_second": 1.934, | |
| "eval_steps_per_second": 0.971, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.27859733978234585, | |
| "grad_norm": 4.718521595001221, | |
| "learning_rate": 8.378375419287099e-05, | |
| "loss": 2.0347, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.2889157597742846, | |
| "grad_norm": 4.244528293609619, | |
| "learning_rate": 8.25432560969328e-05, | |
| "loss": 2.0569, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.2889157597742846, | |
| "eval_loss": 2.052781105041504, | |
| "eval_runtime": 252.644, | |
| "eval_samples_per_second": 0.97, | |
| "eval_steps_per_second": 0.487, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.2992341797662233, | |
| "grad_norm": 5.245513439178467, | |
| "learning_rate": 8.126698441107146e-05, | |
| "loss": 2.1042, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.309552599758162, | |
| "grad_norm": 5.125065326690674, | |
| "learning_rate": 7.995634209323886e-05, | |
| "loss": 2.1062, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.309552599758162, | |
| "eval_loss": 2.035524845123291, | |
| "eval_runtime": 141.6722, | |
| "eval_samples_per_second": 1.729, | |
| "eval_steps_per_second": 0.868, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.3198710197501008, | |
| "grad_norm": 3.9797890186309814, | |
| "learning_rate": 7.861276988374302e-05, | |
| "loss": 2.0977, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.3301894397420395, | |
| "grad_norm": 4.666868686676025, | |
| "learning_rate": 7.723774472149601e-05, | |
| "loss": 2.0522, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.3301894397420395, | |
| "eval_loss": 2.036526679992676, | |
| "eval_runtime": 133.5887, | |
| "eval_samples_per_second": 1.834, | |
| "eval_steps_per_second": 0.921, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.34050785973397824, | |
| "grad_norm": 4.081684589385986, | |
| "learning_rate": 7.583277812046993e-05, | |
| "loss": 2.0314, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.35082627972591696, | |
| "grad_norm": 4.402740955352783, | |
| "learning_rate": 7.439941450814591e-05, | |
| "loss": 2.0631, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.35082627972591696, | |
| "eval_loss": 2.029966115951538, | |
| "eval_runtime": 129.43, | |
| "eval_samples_per_second": 1.893, | |
| "eval_steps_per_second": 0.95, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.3611446997178557, | |
| "grad_norm": 4.357232570648193, | |
| "learning_rate": 7.293922952778239e-05, | |
| "loss": 2.0651, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.37146311970979445, | |
| "grad_norm": 4.258504390716553, | |
| "learning_rate": 7.145382830636924e-05, | |
| "loss": 2.1052, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.37146311970979445, | |
| "eval_loss": 2.040853500366211, | |
| "eval_runtime": 266.0097, | |
| "eval_samples_per_second": 0.921, | |
| "eval_steps_per_second": 0.462, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.3817815397017332, | |
| "grad_norm": 4.097599506378174, | |
| "learning_rate": 6.994484369017143e-05, | |
| "loss": 2.0833, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.3920999596936719, | |
| "grad_norm": 4.530006408691406, | |
| "learning_rate": 6.841393444980177e-05, | |
| "loss": 2.0875, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.3920999596936719, | |
| "eval_loss": 2.045410394668579, | |
| "eval_runtime": 253.392, | |
| "eval_samples_per_second": 0.967, | |
| "eval_steps_per_second": 0.485, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.4024183796856106, | |
| "grad_norm": 4.209381580352783, | |
| "learning_rate": 6.686278345679625e-05, | |
| "loss": 2.1119, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.4127367996775494, | |
| "grad_norm": 5.037073612213135, | |
| "learning_rate": 6.529309583369605e-05, | |
| "loss": 2.0854, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4127367996775494, | |
| "eval_loss": 2.02734112739563, | |
| "eval_runtime": 140.0201, | |
| "eval_samples_per_second": 1.75, | |
| "eval_steps_per_second": 0.878, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4230552196694881, | |
| "grad_norm": 3.838669776916504, | |
| "learning_rate": 6.370659707966967e-05, | |
| "loss": 2.0698, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.43337363966142683, | |
| "grad_norm": 4.199975967407227, | |
| "learning_rate": 6.2105031173736e-05, | |
| "loss": 2.0533, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.43337363966142683, | |
| "eval_loss": 2.0528602600097656, | |
| "eval_runtime": 136.9999, | |
| "eval_samples_per_second": 1.788, | |
| "eval_steps_per_second": 0.898, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.44369205965336556, | |
| "grad_norm": 4.185637950897217, | |
| "learning_rate": 6.049015865767318e-05, | |
| "loss": 2.0949, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.45401047964530433, | |
| "grad_norm": 4.430467128753662, | |
| "learning_rate": 5.88637547007204e-05, | |
| "loss": 2.1096, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.45401047964530433, | |
| "eval_loss": 2.0372908115386963, | |
| "eval_runtime": 126.2447, | |
| "eval_samples_per_second": 1.941, | |
| "eval_steps_per_second": 0.974, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.46432889963724305, | |
| "grad_norm": 4.2635416984558105, | |
| "learning_rate": 5.722760714820057e-05, | |
| "loss": 2.0936, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.4746473196291818, | |
| "grad_norm": 3.9342944622039795, | |
| "learning_rate": 5.5583514556208514e-05, | |
| "loss": 2.0288, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.4746473196291818, | |
| "eval_loss": 2.028881549835205, | |
| "eval_runtime": 251.1061, | |
| "eval_samples_per_second": 0.976, | |
| "eval_steps_per_second": 0.49, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.4849657396211205, | |
| "grad_norm": 4.739011287689209, | |
| "learning_rate": 5.393328421452514e-05, | |
| "loss": 2.0552, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.49528415961305927, | |
| "grad_norm": 4.251812934875488, | |
| "learning_rate": 5.2278730159931076e-05, | |
| "loss": 2.1344, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.49528415961305927, | |
| "eval_loss": 2.037468671798706, | |
| "eval_runtime": 127.6312, | |
| "eval_samples_per_second": 1.92, | |
| "eval_steps_per_second": 0.964, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.505602579604998, | |
| "grad_norm": 4.010369777679443, | |
| "learning_rate": 5.062167118210367e-05, | |
| "loss": 2.1012, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.5159209995969367, | |
| "grad_norm": 4.2648844718933105, | |
| "learning_rate": 4.896392882428901e-05, | |
| "loss": 2.0952, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5159209995969367, | |
| "eval_loss": 2.044463634490967, | |
| "eval_runtime": 144.9327, | |
| "eval_samples_per_second": 1.69, | |
| "eval_steps_per_second": 0.849, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5262394195888754, | |
| "grad_norm": 5.115091800689697, | |
| "learning_rate": 4.730732538094749e-05, | |
| "loss": 2.092, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.5365578395808142, | |
| "grad_norm": 4.243156909942627, | |
| "learning_rate": 4.565368189457313e-05, | |
| "loss": 2.0613, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5365578395808142, | |
| "eval_loss": 2.0373597145080566, | |
| "eval_runtime": 139.7505, | |
| "eval_samples_per_second": 1.753, | |
| "eval_steps_per_second": 0.88, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5468762595727529, | |
| "grad_norm": 4.147082328796387, | |
| "learning_rate": 4.400481615388948e-05, | |
| "loss": 2.0742, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.5571946795646917, | |
| "grad_norm": 4.240519046783447, | |
| "learning_rate": 4.236254069562213e-05, | |
| "loss": 2.0441, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5571946795646917, | |
| "eval_loss": 2.0225021839141846, | |
| "eval_runtime": 140.6705, | |
| "eval_samples_per_second": 1.742, | |
| "eval_steps_per_second": 0.874, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5675130995566304, | |
| "grad_norm": 5.2874650955200195, | |
| "learning_rate": 4.0728660812044536e-05, | |
| "loss": 2.034, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.5778315195485692, | |
| "grad_norm": 4.569270610809326, | |
| "learning_rate": 3.910497256648742e-05, | |
| "loss": 2.0493, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5778315195485692, | |
| "eval_loss": 2.0379703044891357, | |
| "eval_runtime": 245.5153, | |
| "eval_samples_per_second": 0.998, | |
| "eval_steps_per_second": 0.501, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5881499395405079, | |
| "grad_norm": 3.7465648651123047, | |
| "learning_rate": 3.749326081899329e-05, | |
| "loss": 2.0379, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.5984683595324466, | |
| "grad_norm": 4.019279479980469, | |
| "learning_rate": 3.589529726428615e-05, | |
| "loss": 2.0568, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.5984683595324466, | |
| "eval_loss": 2.021915912628174, | |
| "eval_runtime": 134.5206, | |
| "eval_samples_per_second": 1.821, | |
| "eval_steps_per_second": 0.914, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.6087867795243853, | |
| "grad_norm": 3.8277833461761475, | |
| "learning_rate": 3.431283848421347e-05, | |
| "loss": 2.021, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.619105199516324, | |
| "grad_norm": 3.819260358810425, | |
| "learning_rate": 3.274762401680124e-05, | |
| "loss": 2.0477, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.619105199516324, | |
| "eval_loss": 2.0260963439941406, | |
| "eval_runtime": 128.1656, | |
| "eval_samples_per_second": 1.912, | |
| "eval_steps_per_second": 0.96, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6294236195082628, | |
| "grad_norm": 3.959582805633545, | |
| "learning_rate": 3.120137444404442e-05, | |
| "loss": 2.073, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.6397420395002016, | |
| "grad_norm": 4.131293296813965, | |
| "learning_rate": 2.9675789500535328e-05, | |
| "loss": 2.1065, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6397420395002016, | |
| "eval_loss": 2.0310373306274414, | |
| "eval_runtime": 121.8664, | |
| "eval_samples_per_second": 2.01, | |
| "eval_steps_per_second": 1.009, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6500604594921403, | |
| "grad_norm": 3.729912519454956, | |
| "learning_rate": 2.8172546205008683e-05, | |
| "loss": 2.0231, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.660378879484079, | |
| "grad_norm": 3.492910861968994, | |
| "learning_rate": 2.6693297016857188e-05, | |
| "loss": 2.0245, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.660378879484079, | |
| "eval_loss": 2.020824670791626, | |
| "eval_runtime": 138.0864, | |
| "eval_samples_per_second": 1.774, | |
| "eval_steps_per_second": 0.891, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6706972994760177, | |
| "grad_norm": 4.330714225769043, | |
| "learning_rate": 2.523966801964468e-05, | |
| "loss": 2.0436, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.6810157194679565, | |
| "grad_norm": 4.122495174407959, | |
| "learning_rate": 2.3813257133612827e-05, | |
| "loss": 2.1013, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6810157194679565, | |
| "eval_loss": 2.0270016193389893, | |
| "eval_runtime": 254.5885, | |
| "eval_samples_per_second": 0.962, | |
| "eval_steps_per_second": 0.483, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6913341394598952, | |
| "grad_norm": 4.589378833770752, | |
| "learning_rate": 2.2415632359146856e-05, | |
| "loss": 2.0243, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.7016525594518339, | |
| "grad_norm": 3.9164700508117676, | |
| "learning_rate": 2.104833005313131e-05, | |
| "loss": 2.0356, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7016525594518339, | |
| "eval_loss": 2.0204756259918213, | |
| "eval_runtime": 139.155, | |
| "eval_samples_per_second": 1.761, | |
| "eval_steps_per_second": 0.884, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7119709794437726, | |
| "grad_norm": 4.024805545806885, | |
| "learning_rate": 1.971285324008994e-05, | |
| "loss": 2.0727, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 0.7222893994357114, | |
| "grad_norm": 3.9658894538879395, | |
| "learning_rate": 1.84106699599668e-05, | |
| "loss": 2.0815, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7222893994357114, | |
| "eval_loss": 2.011732339859009, | |
| "eval_runtime": 135.732, | |
| "eval_samples_per_second": 1.805, | |
| "eval_steps_per_second": 0.906, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7326078194276502, | |
| "grad_norm": 4.008228778839111, | |
| "learning_rate": 1.7143211654364762e-05, | |
| "loss": 2.0872, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 0.7429262394195889, | |
| "grad_norm": 4.043550968170166, | |
| "learning_rate": 1.5911871593014837e-05, | |
| "loss": 2.0898, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7429262394195889, | |
| "eval_loss": 2.0174756050109863, | |
| "eval_runtime": 134.8529, | |
| "eval_samples_per_second": 1.817, | |
| "eval_steps_per_second": 0.912, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7532446594115276, | |
| "grad_norm": 4.166672706604004, | |
| "learning_rate": 1.4718003342206722e-05, | |
| "loss": 2.0861, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 0.7635630794034663, | |
| "grad_norm": 4.1882100105285645, | |
| "learning_rate": 1.3562919276863844e-05, | |
| "loss": 2.0529, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7635630794034663, | |
| "eval_loss": 2.017113208770752, | |
| "eval_runtime": 126.4791, | |
| "eval_samples_per_second": 1.937, | |
| "eval_steps_per_second": 0.972, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7738814993954051, | |
| "grad_norm": 4.4058709144592285, | |
| "learning_rate": 1.2447889137898293e-05, | |
| "loss": 2.0136, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7841999193873438, | |
| "grad_norm": 4.662993431091309, | |
| "learning_rate": 1.1374138636432053e-05, | |
| "loss": 2.0281, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7841999193873438, | |
| "eval_loss": 2.0133650302886963, | |
| "eval_runtime": 244.3912, | |
| "eval_samples_per_second": 1.002, | |
| "eval_steps_per_second": 0.503, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7945183393792825, | |
| "grad_norm": 4.017709732055664, | |
| "learning_rate": 1.0342848106418368e-05, | |
| "loss": 2.0556, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.8048367593712212, | |
| "grad_norm": 4.138412952423096, | |
| "learning_rate": 9.35515120714447e-06, | |
| "loss": 2.0473, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8048367593712212, | |
| "eval_loss": 2.01501727104187, | |
| "eval_runtime": 142.6675, | |
| "eval_samples_per_second": 1.717, | |
| "eval_steps_per_second": 0.862, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8151551793631601, | |
| "grad_norm": 4.134696006774902, | |
| "learning_rate": 8.41213367704224e-06, | |
| "loss": 2.0193, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 0.8254735993550988, | |
| "grad_norm": 3.4751768112182617, | |
| "learning_rate": 7.51483214017637e-06, | |
| "loss": 2.0315, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8254735993550988, | |
| "eval_loss": 2.008768320083618, | |
| "eval_runtime": 122.1967, | |
| "eval_samples_per_second": 2.005, | |
| "eval_steps_per_second": 1.007, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8357920193470375, | |
| "grad_norm": 4.169270038604736, | |
| "learning_rate": 6.664232966721995e-06, | |
| "loss": 2.1053, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 0.8461104393389762, | |
| "grad_norm": 3.8954155445098877, | |
| "learning_rate": 5.8612711886848196e-06, | |
| "loss": 2.0215, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8461104393389762, | |
| "eval_loss": 2.0070595741271973, | |
| "eval_runtime": 130.4734, | |
| "eval_samples_per_second": 1.878, | |
| "eval_steps_per_second": 0.943, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.856428859330915, | |
| "grad_norm": 4.543677806854248, | |
| "learning_rate": 5.106829472055202e-06, | |
| "loss": 2.046, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 0.8667472793228537, | |
| "grad_norm": 3.7477471828460693, | |
| "learning_rate": 4.401737146526219e-06, | |
| "loss": 2.0003, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8667472793228537, | |
| "eval_loss": 2.0092661380767822, | |
| "eval_runtime": 263.0794, | |
| "eval_samples_per_second": 0.931, | |
| "eval_steps_per_second": 0.468, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8770656993147924, | |
| "grad_norm": 3.9242446422576904, | |
| "learning_rate": 3.7467692938425057e-06, | |
| "loss": 2.0887, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.8873841193067311, | |
| "grad_norm": 4.085505962371826, | |
| "learning_rate": 3.142645895781715e-06, | |
| "loss": 2.0561, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8873841193067311, | |
| "eval_loss": 2.013634443283081, | |
| "eval_runtime": 131.734, | |
| "eval_samples_per_second": 1.86, | |
| "eval_steps_per_second": 0.934, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8977025392986699, | |
| "grad_norm": 3.602693557739258, | |
| "learning_rate": 2.5900310427053044e-06, | |
| "loss": 2.0014, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.9080209592906087, | |
| "grad_norm": 4.137627124786377, | |
| "learning_rate": 2.089532203548794e-06, | |
| "loss": 2.0407, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9080209592906087, | |
| "eval_loss": 2.013155460357666, | |
| "eval_runtime": 140.8366, | |
| "eval_samples_per_second": 1.74, | |
| "eval_steps_per_second": 0.873, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9183393792825474, | |
| "grad_norm": 4.600844860076904, | |
| "learning_rate": 1.6416995580537664e-06, | |
| "loss": 2.0908, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 0.9286577992744861, | |
| "grad_norm": 3.9122657775878906, | |
| "learning_rate": 1.247025391975698e-06, | |
| "loss": 2.0257, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9286577992744861, | |
| "eval_loss": 2.0104849338531494, | |
| "eval_runtime": 131.7523, | |
| "eval_samples_per_second": 1.86, | |
| "eval_steps_per_second": 0.934, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9389762192664248, | |
| "grad_norm": 3.483215093612671, | |
| "learning_rate": 9.059435559326257e-07, | |
| "loss": 2.0517, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.9492946392583635, | |
| "grad_norm": 3.709639072418213, | |
| "learning_rate": 6.188289884893062e-07, | |
| "loss": 2.0294, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9492946392583635, | |
| "eval_loss": 2.0089664459228516, | |
| "eval_runtime": 143.1227, | |
| "eval_samples_per_second": 1.712, | |
| "eval_steps_per_second": 0.859, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9596130592503023, | |
| "grad_norm": 3.766313314437866, | |
| "learning_rate": 3.8599730400115107e-07, | |
| "loss": 2.0648, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 0.969931479242241, | |
| "grad_norm": 3.7077174186706543, | |
| "learning_rate": 2.0770444567118075e-07, | |
| "loss": 2.0321, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.969931479242241, | |
| "eval_loss": 2.0088565349578857, | |
| "eval_runtime": 248.9022, | |
| "eval_samples_per_second": 0.984, | |
| "eval_steps_per_second": 0.494, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9802498992341797, | |
| "grad_norm": 3.656935930252075, | |
| "learning_rate": 8.414640420116305e-08, | |
| "loss": 2.0558, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.9905683192261185, | |
| "grad_norm": 3.686316967010498, | |
| "learning_rate": 1.5459002346324135e-08, | |
| "loss": 2.0516, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9905683192261185, | |
| "eval_loss": 2.009140729904175, | |
| "eval_runtime": 135.7573, | |
| "eval_samples_per_second": 1.805, | |
| "eval_steps_per_second": 0.906, | |
| "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": 5.875183525374198e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |