| { | |
| "best_metric": 0.8090225458145142, | |
| "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l58-l/checkpoint-31500", | |
| "epoch": 3.359641638225256, | |
| "eval_steps": 500, | |
| "global_step": 31500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05332764505119454, | |
| "grad_norm": 217.55360412597656, | |
| "learning_rate": 4.933340443686007e-08, | |
| "loss": 0.4667, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.05332764505119454, | |
| "eval_loss": 1.4426143169403076, | |
| "eval_runtime": 61.9901, | |
| "eval_samples_per_second": 15.922, | |
| "eval_steps_per_second": 2.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.10665529010238908, | |
| "grad_norm": 73.73340606689453, | |
| "learning_rate": 4.8666808873720136e-08, | |
| "loss": 0.4532, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.10665529010238908, | |
| "eval_loss": 1.3815597295761108, | |
| "eval_runtime": 62.4624, | |
| "eval_samples_per_second": 15.801, | |
| "eval_steps_per_second": 1.985, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1599829351535836, | |
| "grad_norm": 471.4011535644531, | |
| "learning_rate": 4.80002133105802e-08, | |
| "loss": 0.3749, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.1599829351535836, | |
| "eval_loss": 1.3310539722442627, | |
| "eval_runtime": 63.3918, | |
| "eval_samples_per_second": 15.57, | |
| "eval_steps_per_second": 1.956, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.21331058020477817, | |
| "grad_norm": 11.962693214416504, | |
| "learning_rate": 4.733361774744027e-08, | |
| "loss": 0.336, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.21331058020477817, | |
| "eval_loss": 1.2890639305114746, | |
| "eval_runtime": 63.177, | |
| "eval_samples_per_second": 15.623, | |
| "eval_steps_per_second": 1.963, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.2666382252559727, | |
| "grad_norm": 375.1609191894531, | |
| "learning_rate": 4.666702218430034e-08, | |
| "loss": 0.3585, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.2666382252559727, | |
| "eval_loss": 1.2536433935165405, | |
| "eval_runtime": 63.1512, | |
| "eval_samples_per_second": 15.629, | |
| "eval_steps_per_second": 1.964, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.3199658703071672, | |
| "grad_norm": 0.0022936267778277397, | |
| "learning_rate": 4.600042662116041e-08, | |
| "loss": 0.303, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3199658703071672, | |
| "eval_loss": 1.2202869653701782, | |
| "eval_runtime": 63.5857, | |
| "eval_samples_per_second": 15.522, | |
| "eval_steps_per_second": 1.95, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.37329351535836175, | |
| "grad_norm": 15.805087089538574, | |
| "learning_rate": 4.5333831058020476e-08, | |
| "loss": 0.3242, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.37329351535836175, | |
| "eval_loss": 1.195627212524414, | |
| "eval_runtime": 63.4024, | |
| "eval_samples_per_second": 15.567, | |
| "eval_steps_per_second": 1.956, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.42662116040955633, | |
| "grad_norm": 5.377908229827881, | |
| "learning_rate": 4.4667235494880546e-08, | |
| "loss": 0.2427, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.42662116040955633, | |
| "eval_loss": 1.169384241104126, | |
| "eval_runtime": 63.5082, | |
| "eval_samples_per_second": 15.541, | |
| "eval_steps_per_second": 1.953, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.47994880546075086, | |
| "grad_norm": 150.5334930419922, | |
| "learning_rate": 4.4000639931740615e-08, | |
| "loss": 0.2993, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.47994880546075086, | |
| "eval_loss": 1.145558476448059, | |
| "eval_runtime": 62.3888, | |
| "eval_samples_per_second": 15.82, | |
| "eval_steps_per_second": 1.988, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.5332764505119454, | |
| "grad_norm": 0.12246542423963547, | |
| "learning_rate": 4.333404436860068e-08, | |
| "loss": 0.3183, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5332764505119454, | |
| "eval_loss": 1.1201218366622925, | |
| "eval_runtime": 63.7154, | |
| "eval_samples_per_second": 15.491, | |
| "eval_steps_per_second": 1.946, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.58660409556314, | |
| "grad_norm": 0.3306196630001068, | |
| "learning_rate": 4.266744880546075e-08, | |
| "loss": 0.307, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.58660409556314, | |
| "eval_loss": 1.0982334613800049, | |
| "eval_runtime": 62.4882, | |
| "eval_samples_per_second": 15.795, | |
| "eval_steps_per_second": 1.984, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.6399317406143344, | |
| "grad_norm": 149.38189697265625, | |
| "learning_rate": 4.2000853242320816e-08, | |
| "loss": 0.2638, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6399317406143344, | |
| "eval_loss": 1.0779541730880737, | |
| "eval_runtime": 63.4101, | |
| "eval_samples_per_second": 15.565, | |
| "eval_steps_per_second": 1.956, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.693259385665529, | |
| "grad_norm": 353.0036315917969, | |
| "learning_rate": 4.1334257679180886e-08, | |
| "loss": 0.2226, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.693259385665529, | |
| "eval_loss": 1.0613036155700684, | |
| "eval_runtime": 62.8292, | |
| "eval_samples_per_second": 15.709, | |
| "eval_steps_per_second": 1.974, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.7465870307167235, | |
| "grad_norm": 0.13188917934894562, | |
| "learning_rate": 4.0667662116040955e-08, | |
| "loss": 0.2453, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7465870307167235, | |
| "eval_loss": 1.0443507432937622, | |
| "eval_runtime": 62.7457, | |
| "eval_samples_per_second": 15.73, | |
| "eval_steps_per_second": 1.976, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7999146757679181, | |
| "grad_norm": 59.391639709472656, | |
| "learning_rate": 4.000106655290102e-08, | |
| "loss": 0.272, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.7999146757679181, | |
| "eval_loss": 1.0300519466400146, | |
| "eval_runtime": 62.3719, | |
| "eval_samples_per_second": 15.824, | |
| "eval_steps_per_second": 1.988, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.8532423208191127, | |
| "grad_norm": 495.4797668457031, | |
| "learning_rate": 3.9334470989761093e-08, | |
| "loss": 0.283, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.8532423208191127, | |
| "eval_loss": 1.0166898965835571, | |
| "eval_runtime": 62.0338, | |
| "eval_samples_per_second": 15.911, | |
| "eval_steps_per_second": 1.999, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.9065699658703071, | |
| "grad_norm": 159.0310821533203, | |
| "learning_rate": 3.866787542662116e-08, | |
| "loss": 0.2331, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.9065699658703071, | |
| "eval_loss": 1.003544807434082, | |
| "eval_runtime": 62.9704, | |
| "eval_samples_per_second": 15.674, | |
| "eval_steps_per_second": 1.969, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.9598976109215017, | |
| "grad_norm": 45.40327835083008, | |
| "learning_rate": 3.8001279863481225e-08, | |
| "loss": 0.2362, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.9598976109215017, | |
| "eval_loss": 0.992453396320343, | |
| "eval_runtime": 63.1405, | |
| "eval_samples_per_second": 15.632, | |
| "eval_steps_per_second": 1.964, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.0132252559726962, | |
| "grad_norm": 621.8538818359375, | |
| "learning_rate": 3.7334684300341295e-08, | |
| "loss": 0.2396, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.0132252559726962, | |
| "eval_loss": 0.9829725027084351, | |
| "eval_runtime": 62.3244, | |
| "eval_samples_per_second": 15.836, | |
| "eval_steps_per_second": 1.99, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.0665529010238908, | |
| "grad_norm": 461.60736083984375, | |
| "learning_rate": 3.6668088737201364e-08, | |
| "loss": 0.2013, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.0665529010238908, | |
| "eval_loss": 0.9735796451568604, | |
| "eval_runtime": 63.0763, | |
| "eval_samples_per_second": 15.648, | |
| "eval_steps_per_second": 1.966, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.1198805460750854, | |
| "grad_norm": 7.0309906005859375, | |
| "learning_rate": 3.6001493174061433e-08, | |
| "loss": 0.2082, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.1198805460750854, | |
| "eval_loss": 0.9639086127281189, | |
| "eval_runtime": 63.0238, | |
| "eval_samples_per_second": 15.661, | |
| "eval_steps_per_second": 1.968, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.17320819112628, | |
| "grad_norm": 304.01995849609375, | |
| "learning_rate": 3.5334897610921496e-08, | |
| "loss": 0.2023, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.17320819112628, | |
| "eval_loss": 0.9558469653129578, | |
| "eval_runtime": 62.1866, | |
| "eval_samples_per_second": 15.872, | |
| "eval_steps_per_second": 1.994, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.2265358361774745, | |
| "grad_norm": 0.0017981340643018484, | |
| "learning_rate": 3.4668302047781565e-08, | |
| "loss": 0.2331, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.2265358361774745, | |
| "eval_loss": 0.9465070962905884, | |
| "eval_runtime": 63.1547, | |
| "eval_samples_per_second": 15.628, | |
| "eval_steps_per_second": 1.963, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.2798634812286689, | |
| "grad_norm": 0.8147886395454407, | |
| "learning_rate": 3.4001706484641635e-08, | |
| "loss": 0.1784, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.2798634812286689, | |
| "eval_loss": 0.9392343163490295, | |
| "eval_runtime": 62.9221, | |
| "eval_samples_per_second": 15.686, | |
| "eval_steps_per_second": 1.971, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.3331911262798635, | |
| "grad_norm": 0.01276802271604538, | |
| "learning_rate": 3.3335110921501704e-08, | |
| "loss": 0.1953, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.3331911262798635, | |
| "eval_loss": 0.9316345453262329, | |
| "eval_runtime": 62.2816, | |
| "eval_samples_per_second": 15.847, | |
| "eval_steps_per_second": 1.991, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.386518771331058, | |
| "grad_norm": 0.07341516762971878, | |
| "learning_rate": 3.2668515358361773e-08, | |
| "loss": 0.1867, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.386518771331058, | |
| "eval_loss": 0.9270105957984924, | |
| "eval_runtime": 63.2619, | |
| "eval_samples_per_second": 15.602, | |
| "eval_steps_per_second": 1.96, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.4398464163822526, | |
| "grad_norm": 17.55596923828125, | |
| "learning_rate": 3.200191979522184e-08, | |
| "loss": 0.22, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.4398464163822526, | |
| "eval_loss": 0.919725775718689, | |
| "eval_runtime": 63.3459, | |
| "eval_samples_per_second": 15.581, | |
| "eval_steps_per_second": 1.958, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.493174061433447, | |
| "grad_norm": 3.4780771732330322, | |
| "learning_rate": 3.133532423208191e-08, | |
| "loss": 0.1656, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.493174061433447, | |
| "eval_loss": 0.9148014783859253, | |
| "eval_runtime": 62.257, | |
| "eval_samples_per_second": 15.854, | |
| "eval_steps_per_second": 1.992, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.5465017064846416, | |
| "grad_norm": 158.22390747070312, | |
| "learning_rate": 3.066872866894198e-08, | |
| "loss": 0.1968, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.5465017064846416, | |
| "eval_loss": 0.9096349477767944, | |
| "eval_runtime": 63.3671, | |
| "eval_samples_per_second": 15.576, | |
| "eval_steps_per_second": 1.957, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.5998293515358362, | |
| "grad_norm": 621.3922729492188, | |
| "learning_rate": 3.0002133105802044e-08, | |
| "loss": 0.1676, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.5998293515358362, | |
| "eval_loss": 0.9056846499443054, | |
| "eval_runtime": 63.2369, | |
| "eval_samples_per_second": 15.608, | |
| "eval_steps_per_second": 1.961, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.6531569965870307, | |
| "grad_norm": 26.23398208618164, | |
| "learning_rate": 2.9335537542662113e-08, | |
| "loss": 0.2074, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.6531569965870307, | |
| "eval_loss": 0.8993550539016724, | |
| "eval_runtime": 62.3767, | |
| "eval_samples_per_second": 15.823, | |
| "eval_steps_per_second": 1.988, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.7064846416382253, | |
| "grad_norm": 283.3314514160156, | |
| "learning_rate": 2.8668941979522183e-08, | |
| "loss": 0.1847, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.7064846416382253, | |
| "eval_loss": 0.8954482078552246, | |
| "eval_runtime": 63.2168, | |
| "eval_samples_per_second": 15.613, | |
| "eval_steps_per_second": 1.962, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.75981228668942, | |
| "grad_norm": 12.151030540466309, | |
| "learning_rate": 2.800234641638225e-08, | |
| "loss": 0.1845, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.75981228668942, | |
| "eval_loss": 0.8900338411331177, | |
| "eval_runtime": 63.5581, | |
| "eval_samples_per_second": 15.529, | |
| "eval_steps_per_second": 1.951, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.8131399317406145, | |
| "grad_norm": 0.7503092288970947, | |
| "learning_rate": 2.7335750853242318e-08, | |
| "loss": 0.1721, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.8131399317406145, | |
| "eval_loss": 0.8872589468955994, | |
| "eval_runtime": 62.0282, | |
| "eval_samples_per_second": 15.912, | |
| "eval_steps_per_second": 1.999, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.8664675767918089, | |
| "grad_norm": 180.9355926513672, | |
| "learning_rate": 2.6669155290102384e-08, | |
| "loss": 0.2627, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.8664675767918089, | |
| "eval_loss": 0.8810189962387085, | |
| "eval_runtime": 62.7212, | |
| "eval_samples_per_second": 15.736, | |
| "eval_steps_per_second": 1.977, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.9197952218430034, | |
| "grad_norm": 0.10503419488668442, | |
| "learning_rate": 2.600255972696246e-08, | |
| "loss": 0.1623, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.9197952218430034, | |
| "eval_loss": 0.8773773908615112, | |
| "eval_runtime": 62.9181, | |
| "eval_samples_per_second": 15.687, | |
| "eval_steps_per_second": 1.971, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.9731228668941978, | |
| "grad_norm": 477.76715087890625, | |
| "learning_rate": 2.5335964163822526e-08, | |
| "loss": 0.2162, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.9731228668941978, | |
| "eval_loss": 0.8713328242301941, | |
| "eval_runtime": 62.3477, | |
| "eval_samples_per_second": 15.831, | |
| "eval_steps_per_second": 1.989, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.0264505119453924, | |
| "grad_norm": 222.06802368164062, | |
| "learning_rate": 2.4669368600682595e-08, | |
| "loss": 0.1802, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.0264505119453924, | |
| "eval_loss": 0.8679063320159912, | |
| "eval_runtime": 63.0768, | |
| "eval_samples_per_second": 15.648, | |
| "eval_steps_per_second": 1.966, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.079778156996587, | |
| "grad_norm": 48.63727569580078, | |
| "learning_rate": 2.400277303754266e-08, | |
| "loss": 0.179, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.079778156996587, | |
| "eval_loss": 0.86326664686203, | |
| "eval_runtime": 63.3068, | |
| "eval_samples_per_second": 15.591, | |
| "eval_steps_per_second": 1.959, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.1331058020477816, | |
| "grad_norm": 0.5827552080154419, | |
| "learning_rate": 2.333617747440273e-08, | |
| "loss": 0.1549, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.1331058020477816, | |
| "eval_loss": 0.8606277108192444, | |
| "eval_runtime": 62.3116, | |
| "eval_samples_per_second": 15.84, | |
| "eval_steps_per_second": 1.99, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.186433447098976, | |
| "grad_norm": 0.060927845537662506, | |
| "learning_rate": 2.2669581911262796e-08, | |
| "loss": 0.1742, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.186433447098976, | |
| "eval_loss": 0.8584741353988647, | |
| "eval_runtime": 63.1659, | |
| "eval_samples_per_second": 15.626, | |
| "eval_steps_per_second": 1.963, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.2397610921501707, | |
| "grad_norm": 16.044307708740234, | |
| "learning_rate": 2.2002986348122866e-08, | |
| "loss": 0.1448, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.2397610921501707, | |
| "eval_loss": 0.8546220064163208, | |
| "eval_runtime": 63.0133, | |
| "eval_samples_per_second": 15.663, | |
| "eval_steps_per_second": 1.968, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.2930887372013653, | |
| "grad_norm": 0.07948703318834305, | |
| "learning_rate": 2.1336390784982935e-08, | |
| "loss": 0.2066, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.2930887372013653, | |
| "eval_loss": 0.8512936234474182, | |
| "eval_runtime": 62.3887, | |
| "eval_samples_per_second": 15.82, | |
| "eval_steps_per_second": 1.988, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.34641638225256, | |
| "grad_norm": 0.0013235179940238595, | |
| "learning_rate": 2.0669795221843004e-08, | |
| "loss": 0.1483, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.34641638225256, | |
| "eval_loss": 0.8480631709098816, | |
| "eval_runtime": 63.1614, | |
| "eval_samples_per_second": 15.627, | |
| "eval_steps_per_second": 1.963, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.399744027303754, | |
| "grad_norm": 0.001996780978515744, | |
| "learning_rate": 2.000319965870307e-08, | |
| "loss": 0.1813, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.399744027303754, | |
| "eval_loss": 0.84471195936203, | |
| "eval_runtime": 63.0032, | |
| "eval_samples_per_second": 15.666, | |
| "eval_steps_per_second": 1.968, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.453071672354949, | |
| "grad_norm": 0.24369017779827118, | |
| "learning_rate": 1.933660409556314e-08, | |
| "loss": 0.1617, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.453071672354949, | |
| "eval_loss": 0.8411402702331543, | |
| "eval_runtime": 62.707, | |
| "eval_samples_per_second": 15.74, | |
| "eval_steps_per_second": 1.977, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.506399317406143, | |
| "grad_norm": 9.377041816711426, | |
| "learning_rate": 1.8670008532423206e-08, | |
| "loss": 0.1664, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.506399317406143, | |
| "eval_loss": 0.8393682837486267, | |
| "eval_runtime": 63.143, | |
| "eval_samples_per_second": 15.631, | |
| "eval_steps_per_second": 1.964, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.5597269624573378, | |
| "grad_norm": 118.065673828125, | |
| "learning_rate": 1.800341296928328e-08, | |
| "loss": 0.1786, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.5597269624573378, | |
| "eval_loss": 0.8357825875282288, | |
| "eval_runtime": 63.3183, | |
| "eval_samples_per_second": 15.588, | |
| "eval_steps_per_second": 1.958, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.6130546075085324, | |
| "grad_norm": 3.0652036666870117, | |
| "learning_rate": 1.7336817406143344e-08, | |
| "loss": 0.1465, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.6130546075085324, | |
| "eval_loss": 0.8330481648445129, | |
| "eval_runtime": 63.0565, | |
| "eval_samples_per_second": 15.653, | |
| "eval_steps_per_second": 1.966, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.666382252559727, | |
| "grad_norm": 68.4100570678711, | |
| "learning_rate": 1.6670221843003414e-08, | |
| "loss": 0.1289, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.666382252559727, | |
| "eval_loss": 0.831382691860199, | |
| "eval_runtime": 62.8466, | |
| "eval_samples_per_second": 15.705, | |
| "eval_steps_per_second": 1.973, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.7197098976109215, | |
| "grad_norm": 0.03573409095406532, | |
| "learning_rate": 1.600362627986348e-08, | |
| "loss": 0.1662, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.7197098976109215, | |
| "eval_loss": 0.829609751701355, | |
| "eval_runtime": 63.536, | |
| "eval_samples_per_second": 15.535, | |
| "eval_steps_per_second": 1.952, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.773037542662116, | |
| "grad_norm": 85.1174087524414, | |
| "learning_rate": 1.533703071672355e-08, | |
| "loss": 0.1463, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.773037542662116, | |
| "eval_loss": 0.8262338638305664, | |
| "eval_runtime": 63.1492, | |
| "eval_samples_per_second": 15.63, | |
| "eval_steps_per_second": 1.964, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.8263651877133107, | |
| "grad_norm": 1.988323450088501, | |
| "learning_rate": 1.4670435153583615e-08, | |
| "loss": 0.1471, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.8263651877133107, | |
| "eval_loss": 0.8249030709266663, | |
| "eval_runtime": 63.1063, | |
| "eval_samples_per_second": 15.64, | |
| "eval_steps_per_second": 1.965, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.8796928327645053, | |
| "grad_norm": 14.553828239440918, | |
| "learning_rate": 1.4003839590443686e-08, | |
| "loss": 0.167, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.8796928327645053, | |
| "eval_loss": 0.821887731552124, | |
| "eval_runtime": 63.5261, | |
| "eval_samples_per_second": 15.537, | |
| "eval_steps_per_second": 1.952, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.9330204778157, | |
| "grad_norm": 0.4691038131713867, | |
| "learning_rate": 1.3337244027303754e-08, | |
| "loss": 0.1268, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.9330204778157, | |
| "eval_loss": 0.8203587532043457, | |
| "eval_runtime": 63.3176, | |
| "eval_samples_per_second": 15.588, | |
| "eval_steps_per_second": 1.958, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.986348122866894, | |
| "grad_norm": 327.5917663574219, | |
| "learning_rate": 1.2670648464163821e-08, | |
| "loss": 0.177, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.986348122866894, | |
| "eval_loss": 0.8176783919334412, | |
| "eval_runtime": 63.0051, | |
| "eval_samples_per_second": 15.665, | |
| "eval_steps_per_second": 1.968, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.0396757679180886, | |
| "grad_norm": 1.4327431472338503e-06, | |
| "learning_rate": 1.200405290102389e-08, | |
| "loss": 0.1206, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 3.0396757679180886, | |
| "eval_loss": 0.8165684938430786, | |
| "eval_runtime": 63.4667, | |
| "eval_samples_per_second": 15.551, | |
| "eval_steps_per_second": 1.954, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 3.093003412969283, | |
| "grad_norm": 0.21452489495277405, | |
| "learning_rate": 1.1337457337883958e-08, | |
| "loss": 0.1345, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.093003412969283, | |
| "eval_loss": 0.8155500292778015, | |
| "eval_runtime": 63.0809, | |
| "eval_samples_per_second": 15.647, | |
| "eval_steps_per_second": 1.966, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.1463310580204777, | |
| "grad_norm": 8.160687446594238, | |
| "learning_rate": 1.0670861774744026e-08, | |
| "loss": 0.1907, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 3.1463310580204777, | |
| "eval_loss": 0.8143629431724548, | |
| "eval_runtime": 62.8466, | |
| "eval_samples_per_second": 15.705, | |
| "eval_steps_per_second": 1.973, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 3.1996587030716723, | |
| "grad_norm": 0.04997281730175018, | |
| "learning_rate": 1.0004266211604095e-08, | |
| "loss": 0.1395, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.1996587030716723, | |
| "eval_loss": 0.812623918056488, | |
| "eval_runtime": 63.0553, | |
| "eval_samples_per_second": 15.653, | |
| "eval_steps_per_second": 1.967, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.252986348122867, | |
| "grad_norm": 609.138916015625, | |
| "learning_rate": 9.337670648464163e-09, | |
| "loss": 0.1511, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 3.252986348122867, | |
| "eval_loss": 0.8111589550971985, | |
| "eval_runtime": 63.102, | |
| "eval_samples_per_second": 15.641, | |
| "eval_steps_per_second": 1.965, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 3.3063139931740615, | |
| "grad_norm": 6.588051795959473, | |
| "learning_rate": 8.671075085324232e-09, | |
| "loss": 0.1334, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 3.3063139931740615, | |
| "eval_loss": 0.8101842403411865, | |
| "eval_runtime": 62.7702, | |
| "eval_samples_per_second": 15.724, | |
| "eval_steps_per_second": 1.975, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 3.359641638225256, | |
| "grad_norm": 0.0009110367391258478, | |
| "learning_rate": 8.0044795221843e-09, | |
| "loss": 0.1799, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 3.359641638225256, | |
| "eval_loss": 0.8090225458145142, | |
| "eval_runtime": 63.1535, | |
| "eval_samples_per_second": 15.629, | |
| "eval_steps_per_second": 1.963, | |
| "step": 31500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 37504, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.133310758139252e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |