| { | |
| "best_metric": 2.015996217727661, | |
| "best_model_checkpoint": "/home/datta0/models/lora_final/Mistral-7B-v0.3_pct_reverse_r16/checkpoint-368", | |
| "epoch": 0.9905683192261185, | |
| "eval_steps": 8, | |
| "global_step": 384, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0025796049979846837, | |
| "grad_norm": 16.794553756713867, | |
| "learning_rate": 1.25e-05, | |
| "loss": 2.1524, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010318419991938735, | |
| "grad_norm": 10.606828689575195, | |
| "learning_rate": 5e-05, | |
| "loss": 2.046, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.02063683998387747, | |
| "grad_norm": 10.568798065185547, | |
| "learning_rate": 0.0001, | |
| "loss": 1.9648, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.02063683998387747, | |
| "eval_loss": 2.0392367839813232, | |
| "eval_runtime": 46.0067, | |
| "eval_samples_per_second": 5.325, | |
| "eval_steps_per_second": 2.674, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.030955259975816204, | |
| "grad_norm": 29.913541793823242, | |
| "learning_rate": 9.997251843068762e-05, | |
| "loss": 2.0181, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04127367996775494, | |
| "grad_norm": 20.060449600219727, | |
| "learning_rate": 9.989010393221656e-05, | |
| "loss": 2.0599, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04127367996775494, | |
| "eval_loss": 2.0531184673309326, | |
| "eval_runtime": 45.5707, | |
| "eval_samples_per_second": 5.376, | |
| "eval_steps_per_second": 2.699, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.051592099959693674, | |
| "grad_norm": 42.19817352294922, | |
| "learning_rate": 9.97528470997769e-05, | |
| "loss": 2.1649, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06191051995163241, | |
| "grad_norm": 53.00956344604492, | |
| "learning_rate": 9.956089881469482e-05, | |
| "loss": 2.1274, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06191051995163241, | |
| "eval_loss": 2.057147741317749, | |
| "eval_runtime": 46.5266, | |
| "eval_samples_per_second": 5.266, | |
| "eval_steps_per_second": 2.644, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07222893994357114, | |
| "grad_norm": 36.37834548950195, | |
| "learning_rate": 9.931447007857432e-05, | |
| "loss": 2.084, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.08254735993550988, | |
| "grad_norm": 26.134798049926758, | |
| "learning_rate": 9.901383178135113e-05, | |
| "loss": 2.0718, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08254735993550988, | |
| "eval_loss": 2.0472965240478516, | |
| "eval_runtime": 80.6576, | |
| "eval_samples_per_second": 3.038, | |
| "eval_steps_per_second": 1.525, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.09286577992744861, | |
| "grad_norm": 15.81359577178955, | |
| "learning_rate": 9.865931440351337e-05, | |
| "loss": 2.0784, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10318419991938735, | |
| "grad_norm": 20.870193481445312, | |
| "learning_rate": 9.825130765281668e-05, | |
| "loss": 2.0646, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10318419991938735, | |
| "eval_loss": 2.041971445083618, | |
| "eval_runtime": 46.7629, | |
| "eval_samples_per_second": 5.239, | |
| "eval_steps_per_second": 2.63, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11350261991132608, | |
| "grad_norm": 18.623149871826172, | |
| "learning_rate": 9.779026003589304e-05, | |
| "loss": 2.0528, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.12382103990326482, | |
| "grad_norm": 13.586410522460938, | |
| "learning_rate": 9.727667836522407e-05, | |
| "loss": 2.0883, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12382103990326482, | |
| "eval_loss": 2.046041488647461, | |
| "eval_runtime": 47.9414, | |
| "eval_samples_per_second": 5.11, | |
| "eval_steps_per_second": 2.566, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.13413945989520354, | |
| "grad_norm": 14.266491889953613, | |
| "learning_rate": 9.6711127202021e-05, | |
| "loss": 2.0503, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.1444578798871423, | |
| "grad_norm": 16.555265426635742, | |
| "learning_rate": 9.609422823562345e-05, | |
| "loss": 2.0611, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.1444578798871423, | |
| "eval_loss": 2.0496585369110107, | |
| "eval_runtime": 46.5878, | |
| "eval_samples_per_second": 5.259, | |
| "eval_steps_per_second": 2.64, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.154776299879081, | |
| "grad_norm": 16.17994499206543, | |
| "learning_rate": 9.542665960009959e-05, | |
| "loss": 2.082, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.16509471987101976, | |
| "grad_norm": 16.075532913208008, | |
| "learning_rate": 9.470915512879852e-05, | |
| "loss": 2.0841, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.16509471987101976, | |
| "eval_loss": 2.053602933883667, | |
| "eval_runtime": 147.0784, | |
| "eval_samples_per_second": 1.666, | |
| "eval_steps_per_second": 0.836, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.17541313986295848, | |
| "grad_norm": 20.747583389282227, | |
| "learning_rate": 9.394250354767467e-05, | |
| "loss": 2.1047, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.18573155985489723, | |
| "grad_norm": 20.321056365966797, | |
| "learning_rate": 9.312754760827061e-05, | |
| "loss": 2.0695, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.18573155985489723, | |
| "eval_loss": 2.0687732696533203, | |
| "eval_runtime": 276.898, | |
| "eval_samples_per_second": 0.885, | |
| "eval_steps_per_second": 0.444, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.19604997984683595, | |
| "grad_norm": 21.839866638183594, | |
| "learning_rate": 9.226518316131176e-05, | |
| "loss": 2.1056, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.2063683998387747, | |
| "grad_norm": 16.164918899536133, | |
| "learning_rate": 9.1356358171931e-05, | |
| "loss": 2.0696, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2063683998387747, | |
| "eval_loss": 2.079158067703247, | |
| "eval_runtime": 138.6965, | |
| "eval_samples_per_second": 1.766, | |
| "eval_steps_per_second": 0.887, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.21668681983071342, | |
| "grad_norm": 16.828935623168945, | |
| "learning_rate": 9.040207167760586e-05, | |
| "loss": 2.0932, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.22700523982265217, | |
| "grad_norm": 24.9482479095459, | |
| "learning_rate": 8.940337268995385e-05, | |
| "loss": 2.1315, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.22700523982265217, | |
| "eval_loss": 2.089974880218506, | |
| "eval_runtime": 145.0596, | |
| "eval_samples_per_second": 1.689, | |
| "eval_steps_per_second": 0.848, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.2373236598145909, | |
| "grad_norm": 14.4788179397583, | |
| "learning_rate": 8.836135904159302e-05, | |
| "loss": 2.1236, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.24764207980652964, | |
| "grad_norm": 14.118853569030762, | |
| "learning_rate": 8.727717617933544e-05, | |
| "loss": 2.1466, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.24764207980652964, | |
| "eval_loss": 2.0826759338378906, | |
| "eval_runtime": 129.8085, | |
| "eval_samples_per_second": 1.887, | |
| "eval_steps_per_second": 0.948, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.25796049979846836, | |
| "grad_norm": 16.814769744873047, | |
| "learning_rate": 8.615201590504017e-05, | |
| "loss": 2.1325, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2682789197904071, | |
| "grad_norm": 13.403229713439941, | |
| "learning_rate": 8.498711506550983e-05, | |
| "loss": 2.1575, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.2682789197904071, | |
| "eval_loss": 2.082623243331909, | |
| "eval_runtime": 132.6815, | |
| "eval_samples_per_second": 1.847, | |
| "eval_steps_per_second": 0.927, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.27859733978234585, | |
| "grad_norm": 19.730573654174805, | |
| "learning_rate": 8.378375419287099e-05, | |
| "loss": 2.0873, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.2889157597742846, | |
| "grad_norm": 19.85085678100586, | |
| "learning_rate": 8.25432560969328e-05, | |
| "loss": 2.0925, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.2889157597742846, | |
| "eval_loss": 2.086350440979004, | |
| "eval_runtime": 271.6641, | |
| "eval_samples_per_second": 0.902, | |
| "eval_steps_per_second": 0.453, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.2992341797662233, | |
| "grad_norm": 14.944377899169922, | |
| "learning_rate": 8.126698441107146e-05, | |
| "loss": 2.1405, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.309552599758162, | |
| "grad_norm": 16.105512619018555, | |
| "learning_rate": 7.995634209323886e-05, | |
| "loss": 2.1647, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.309552599758162, | |
| "eval_loss": 2.0815000534057617, | |
| "eval_runtime": 133.9294, | |
| "eval_samples_per_second": 1.829, | |
| "eval_steps_per_second": 0.918, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.3198710197501008, | |
| "grad_norm": 17.4757080078125, | |
| "learning_rate": 7.861276988374302e-05, | |
| "loss": 2.1381, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.3301894397420395, | |
| "grad_norm": 14.932345390319824, | |
| "learning_rate": 7.723774472149601e-05, | |
| "loss": 2.1018, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.3301894397420395, | |
| "eval_loss": 2.0882339477539062, | |
| "eval_runtime": 139.4099, | |
| "eval_samples_per_second": 1.757, | |
| "eval_steps_per_second": 0.882, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.34050785973397824, | |
| "grad_norm": 14.54583740234375, | |
| "learning_rate": 7.583277812046993e-05, | |
| "loss": 2.0851, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.35082627972591696, | |
| "grad_norm": 18.295686721801758, | |
| "learning_rate": 7.439941450814591e-05, | |
| "loss": 2.1062, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.35082627972591696, | |
| "eval_loss": 2.0903666019439697, | |
| "eval_runtime": 138.7624, | |
| "eval_samples_per_second": 1.766, | |
| "eval_steps_per_second": 0.886, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.3611446997178557, | |
| "grad_norm": 14.11962890625, | |
| "learning_rate": 7.293922952778239e-05, | |
| "loss": 2.1298, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.37146311970979445, | |
| "grad_norm": 13.146790504455566, | |
| "learning_rate": 7.145382830636924e-05, | |
| "loss": 2.1596, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.37146311970979445, | |
| "eval_loss": 2.0847134590148926, | |
| "eval_runtime": 271.2553, | |
| "eval_samples_per_second": 0.903, | |
| "eval_steps_per_second": 0.453, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.3817815397017332, | |
| "grad_norm": 14.248947143554688, | |
| "learning_rate": 6.994484369017143e-05, | |
| "loss": 2.1231, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.3920999596936719, | |
| "grad_norm": 15.606071472167969, | |
| "learning_rate": 6.841393444980177e-05, | |
| "loss": 2.1473, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.3920999596936719, | |
| "eval_loss": 2.0933380126953125, | |
| "eval_runtime": 276.868, | |
| "eval_samples_per_second": 0.885, | |
| "eval_steps_per_second": 0.444, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.4024183796856106, | |
| "grad_norm": 14.0585298538208, | |
| "learning_rate": 6.686278345679625e-05, | |
| "loss": 2.1495, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.4127367996775494, | |
| "grad_norm": 17.74847412109375, | |
| "learning_rate": 6.529309583369605e-05, | |
| "loss": 2.1388, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4127367996775494, | |
| "eval_loss": 2.088761568069458, | |
| "eval_runtime": 130.7698, | |
| "eval_samples_per_second": 1.874, | |
| "eval_steps_per_second": 0.941, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4230552196694881, | |
| "grad_norm": 14.887138366699219, | |
| "learning_rate": 6.370659707966967e-05, | |
| "loss": 2.1268, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.43337363966142683, | |
| "grad_norm": 13.487187385559082, | |
| "learning_rate": 6.2105031173736e-05, | |
| "loss": 2.093, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.43337363966142683, | |
| "eval_loss": 2.088651657104492, | |
| "eval_runtime": 139.173, | |
| "eval_samples_per_second": 1.76, | |
| "eval_steps_per_second": 0.884, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.44369205965336556, | |
| "grad_norm": 12.735310554504395, | |
| "learning_rate": 6.049015865767318e-05, | |
| "loss": 2.1416, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.45401047964530433, | |
| "grad_norm": 15.59807014465332, | |
| "learning_rate": 5.88637547007204e-05, | |
| "loss": 2.1704, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.45401047964530433, | |
| "eval_loss": 2.093348979949951, | |
| "eval_runtime": 141.3221, | |
| "eval_samples_per_second": 1.734, | |
| "eval_steps_per_second": 0.87, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.46432889963724305, | |
| "grad_norm": 14.380990028381348, | |
| "learning_rate": 5.722760714820057e-05, | |
| "loss": 2.1412, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.4746473196291818, | |
| "grad_norm": 12.070490837097168, | |
| "learning_rate": 5.5583514556208514e-05, | |
| "loss": 2.0697, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.4746473196291818, | |
| "eval_loss": 2.0778870582580566, | |
| "eval_runtime": 245.7592, | |
| "eval_samples_per_second": 0.997, | |
| "eval_steps_per_second": 0.5, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.4849657396211205, | |
| "grad_norm": 13.802606582641602, | |
| "learning_rate": 5.393328421452514e-05, | |
| "loss": 2.1088, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.49528415961305927, | |
| "grad_norm": 13.323396682739258, | |
| "learning_rate": 5.2278730159931076e-05, | |
| "loss": 2.1725, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.49528415961305927, | |
| "eval_loss": 2.071396589279175, | |
| "eval_runtime": 139.8299, | |
| "eval_samples_per_second": 1.752, | |
| "eval_steps_per_second": 0.88, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.505602579604998, | |
| "grad_norm": 12.754402160644531, | |
| "learning_rate": 5.062167118210367e-05, | |
| "loss": 2.1362, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.5159209995969367, | |
| "grad_norm": 16.00323486328125, | |
| "learning_rate": 4.896392882428901e-05, | |
| "loss": 2.1339, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5159209995969367, | |
| "eval_loss": 2.069511890411377, | |
| "eval_runtime": 139.0327, | |
| "eval_samples_per_second": 1.762, | |
| "eval_steps_per_second": 0.885, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5262394195888754, | |
| "grad_norm": 13.225739479064941, | |
| "learning_rate": 4.730732538094749e-05, | |
| "loss": 2.1241, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.5365578395808142, | |
| "grad_norm": 14.486602783203125, | |
| "learning_rate": 4.565368189457313e-05, | |
| "loss": 2.106, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5365578395808142, | |
| "eval_loss": 2.063960075378418, | |
| "eval_runtime": 141.0631, | |
| "eval_samples_per_second": 1.737, | |
| "eval_steps_per_second": 0.872, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5468762595727529, | |
| "grad_norm": 12.7869234085083, | |
| "learning_rate": 4.400481615388948e-05, | |
| "loss": 2.1111, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.5571946795646917, | |
| "grad_norm": 13.145895957946777, | |
| "learning_rate": 4.236254069562213e-05, | |
| "loss": 2.0857, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5571946795646917, | |
| "eval_loss": 2.0791518688201904, | |
| "eval_runtime": 143.4311, | |
| "eval_samples_per_second": 1.708, | |
| "eval_steps_per_second": 0.858, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5675130995566304, | |
| "grad_norm": 12.685562133789062, | |
| "learning_rate": 4.0728660812044536e-05, | |
| "loss": 2.0832, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.5778315195485692, | |
| "grad_norm": 13.919960975646973, | |
| "learning_rate": 3.910497256648742e-05, | |
| "loss": 2.0751, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5778315195485692, | |
| "eval_loss": 2.0658464431762695, | |
| "eval_runtime": 269.6696, | |
| "eval_samples_per_second": 0.909, | |
| "eval_steps_per_second": 0.456, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5881499395405079, | |
| "grad_norm": 13.588690757751465, | |
| "learning_rate": 3.749326081899329e-05, | |
| "loss": 2.0779, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.5984683595324466, | |
| "grad_norm": 12.779603004455566, | |
| "learning_rate": 3.589529726428615e-05, | |
| "loss": 2.0987, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.5984683595324466, | |
| "eval_loss": 2.0659260749816895, | |
| "eval_runtime": 142.9536, | |
| "eval_samples_per_second": 1.714, | |
| "eval_steps_per_second": 0.86, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.6087867795243853, | |
| "grad_norm": 14.055486679077148, | |
| "learning_rate": 3.431283848421347e-05, | |
| "loss": 2.0618, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.619105199516324, | |
| "grad_norm": 12.830402374267578, | |
| "learning_rate": 3.274762401680124e-05, | |
| "loss": 2.0817, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.619105199516324, | |
| "eval_loss": 2.0628087520599365, | |
| "eval_runtime": 144.7768, | |
| "eval_samples_per_second": 1.692, | |
| "eval_steps_per_second": 0.85, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6294236195082628, | |
| "grad_norm": 11.332627296447754, | |
| "learning_rate": 3.120137444404442e-05, | |
| "loss": 2.1072, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.6397420395002016, | |
| "grad_norm": 11.619379997253418, | |
| "learning_rate": 2.9675789500535328e-05, | |
| "loss": 2.1341, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6397420395002016, | |
| "eval_loss": 2.05637526512146, | |
| "eval_runtime": 134.9288, | |
| "eval_samples_per_second": 1.816, | |
| "eval_steps_per_second": 0.912, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6500604594921403, | |
| "grad_norm": 11.539495468139648, | |
| "learning_rate": 2.8172546205008683e-05, | |
| "loss": 2.0623, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.660378879484079, | |
| "grad_norm": 10.27650260925293, | |
| "learning_rate": 2.6693297016857188e-05, | |
| "loss": 2.0567, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.660378879484079, | |
| "eval_loss": 2.051720142364502, | |
| "eval_runtime": 132.5711, | |
| "eval_samples_per_second": 1.848, | |
| "eval_steps_per_second": 0.928, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6706972994760177, | |
| "grad_norm": 10.966778755187988, | |
| "learning_rate": 2.523966801964468e-05, | |
| "loss": 2.0717, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.6810157194679565, | |
| "grad_norm": 10.349825859069824, | |
| "learning_rate": 2.3813257133612827e-05, | |
| "loss": 2.1246, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6810157194679565, | |
| "eval_loss": 2.045745611190796, | |
| "eval_runtime": 269.5914, | |
| "eval_samples_per_second": 0.909, | |
| "eval_steps_per_second": 0.456, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6913341394598952, | |
| "grad_norm": 10.632711410522461, | |
| "learning_rate": 2.2415632359146856e-05, | |
| "loss": 2.0458, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.7016525594518339, | |
| "grad_norm": 9.394647598266602, | |
| "learning_rate": 2.104833005313131e-05, | |
| "loss": 2.0623, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7016525594518339, | |
| "eval_loss": 2.0422868728637695, | |
| "eval_runtime": 131.0567, | |
| "eval_samples_per_second": 1.869, | |
| "eval_steps_per_second": 0.939, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7119709794437726, | |
| "grad_norm": 11.318469047546387, | |
| "learning_rate": 1.971285324008994e-05, | |
| "loss": 2.0952, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 0.7222893994357114, | |
| "grad_norm": 10.162789344787598, | |
| "learning_rate": 1.84106699599668e-05, | |
| "loss": 2.1106, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7222893994357114, | |
| "eval_loss": 2.0368635654449463, | |
| "eval_runtime": 125.1131, | |
| "eval_samples_per_second": 1.958, | |
| "eval_steps_per_second": 0.983, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7326078194276502, | |
| "grad_norm": 11.464627265930176, | |
| "learning_rate": 1.7143211654364762e-05, | |
| "loss": 2.112, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 0.7429262394195889, | |
| "grad_norm": 12.67012882232666, | |
| "learning_rate": 1.5911871593014837e-05, | |
| "loss": 2.1094, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7429262394195889, | |
| "eval_loss": 2.0374763011932373, | |
| "eval_runtime": 135.9513, | |
| "eval_samples_per_second": 1.802, | |
| "eval_steps_per_second": 0.905, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7532446594115276, | |
| "grad_norm": 11.443336486816406, | |
| "learning_rate": 1.4718003342206722e-05, | |
| "loss": 2.1089, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 0.7635630794034663, | |
| "grad_norm": 11.790454864501953, | |
| "learning_rate": 1.3562919276863844e-05, | |
| "loss": 2.0678, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7635630794034663, | |
| "eval_loss": 2.032977819442749, | |
| "eval_runtime": 145.1862, | |
| "eval_samples_per_second": 1.687, | |
| "eval_steps_per_second": 0.847, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7738814993954051, | |
| "grad_norm": 11.48520565032959, | |
| "learning_rate": 1.2447889137898293e-05, | |
| "loss": 2.0354, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7841999193873438, | |
| "grad_norm": 10.889254570007324, | |
| "learning_rate": 1.1374138636432053e-05, | |
| "loss": 2.0521, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7841999193873438, | |
| "eval_loss": 2.032576560974121, | |
| "eval_runtime": 266.89, | |
| "eval_samples_per_second": 0.918, | |
| "eval_steps_per_second": 0.461, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7945183393792825, | |
| "grad_norm": 11.634953498840332, | |
| "learning_rate": 1.0342848106418368e-05, | |
| "loss": 2.0765, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.8048367593712212, | |
| "grad_norm": 10.742921829223633, | |
| "learning_rate": 9.35515120714447e-06, | |
| "loss": 2.0594, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8048367593712212, | |
| "eval_loss": 2.0241119861602783, | |
| "eval_runtime": 145.6455, | |
| "eval_samples_per_second": 1.682, | |
| "eval_steps_per_second": 0.845, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8151551793631601, | |
| "grad_norm": 9.77017593383789, | |
| "learning_rate": 8.41213367704224e-06, | |
| "loss": 2.0305, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 0.8254735993550988, | |
| "grad_norm": 9.984450340270996, | |
| "learning_rate": 7.51483214017637e-06, | |
| "loss": 2.051, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8254735993550988, | |
| "eval_loss": 2.0207557678222656, | |
| "eval_runtime": 141.4793, | |
| "eval_samples_per_second": 1.732, | |
| "eval_steps_per_second": 0.869, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8357920193470375, | |
| "grad_norm": 10.030397415161133, | |
| "learning_rate": 6.664232966721995e-06, | |
| "loss": 2.1179, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 0.8461104393389762, | |
| "grad_norm": 9.973139762878418, | |
| "learning_rate": 5.8612711886848196e-06, | |
| "loss": 2.0392, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8461104393389762, | |
| "eval_loss": 2.020094156265259, | |
| "eval_runtime": 136.0416, | |
| "eval_samples_per_second": 1.801, | |
| "eval_steps_per_second": 0.904, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.856428859330915, | |
| "grad_norm": 9.772870063781738, | |
| "learning_rate": 5.106829472055202e-06, | |
| "loss": 2.0651, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 0.8667472793228537, | |
| "grad_norm": 10.798050880432129, | |
| "learning_rate": 4.401737146526219e-06, | |
| "loss": 2.0143, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8667472793228537, | |
| "eval_loss": 2.0207338333129883, | |
| "eval_runtime": 254.8265, | |
| "eval_samples_per_second": 0.961, | |
| "eval_steps_per_second": 0.483, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8770656993147924, | |
| "grad_norm": 9.898650169372559, | |
| "learning_rate": 3.7467692938425057e-06, | |
| "loss": 2.0972, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.8873841193067311, | |
| "grad_norm": 9.767654418945312, | |
| "learning_rate": 3.142645895781715e-06, | |
| "loss": 2.0678, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8873841193067311, | |
| "eval_loss": 2.0221924781799316, | |
| "eval_runtime": 135.0488, | |
| "eval_samples_per_second": 1.814, | |
| "eval_steps_per_second": 0.911, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8977025392986699, | |
| "grad_norm": 8.583464622497559, | |
| "learning_rate": 2.5900310427053044e-06, | |
| "loss": 2.0127, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.9080209592906087, | |
| "grad_norm": 10.855988502502441, | |
| "learning_rate": 2.089532203548794e-06, | |
| "loss": 2.0473, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9080209592906087, | |
| "eval_loss": 2.0186901092529297, | |
| "eval_runtime": 141.8755, | |
| "eval_samples_per_second": 1.727, | |
| "eval_steps_per_second": 0.867, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9183393792825474, | |
| "grad_norm": 10.131417274475098, | |
| "learning_rate": 1.6416995580537664e-06, | |
| "loss": 2.0994, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 0.9286577992744861, | |
| "grad_norm": 9.968809127807617, | |
| "learning_rate": 1.247025391975698e-06, | |
| "loss": 2.0324, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9286577992744861, | |
| "eval_loss": 2.016472101211548, | |
| "eval_runtime": 135.0646, | |
| "eval_samples_per_second": 1.814, | |
| "eval_steps_per_second": 0.911, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9389762192664248, | |
| "grad_norm": 9.20028018951416, | |
| "learning_rate": 9.059435559326257e-07, | |
| "loss": 2.0579, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.9492946392583635, | |
| "grad_norm": 9.900856971740723, | |
| "learning_rate": 6.188289884893062e-07, | |
| "loss": 2.0404, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9492946392583635, | |
| "eval_loss": 2.015996217727661, | |
| "eval_runtime": 135.5315, | |
| "eval_samples_per_second": 1.808, | |
| "eval_steps_per_second": 0.908, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9596130592503023, | |
| "grad_norm": 10.436941146850586, | |
| "learning_rate": 3.8599730400115107e-07, | |
| "loss": 2.0711, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 0.969931479242241, | |
| "grad_norm": 10.571781158447266, | |
| "learning_rate": 2.0770444567118075e-07, | |
| "loss": 2.0426, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.969931479242241, | |
| "eval_loss": 2.0162649154663086, | |
| "eval_runtime": 275.4389, | |
| "eval_samples_per_second": 0.889, | |
| "eval_steps_per_second": 0.447, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9802498992341797, | |
| "grad_norm": 8.854607582092285, | |
| "learning_rate": 8.414640420116305e-08, | |
| "loss": 2.0661, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.9905683192261185, | |
| "grad_norm": 9.778440475463867, | |
| "learning_rate": 1.5459002346324135e-08, | |
| "loss": 2.0635, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9905683192261185, | |
| "eval_loss": 2.0161848068237305, | |
| "eval_runtime": 140.5211, | |
| "eval_samples_per_second": 1.744, | |
| "eval_steps_per_second": 0.875, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.83492814456488e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |