| { | |
| "best_metric": 0.7003, | |
| "best_model_checkpoint": "runs/legis-tucano-2b-valid/checkpoint-710", | |
| "epoch": 0.9994065281899109, | |
| "eval_steps": 10, | |
| "global_step": 842, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.011869436201780416, | |
| "grad_norm": 1.3277966052420052, | |
| "learning_rate": 2.3529411764705884e-05, | |
| "loss": 1.3563, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.011869436201780416, | |
| "eval_loss": 1.2333489656448364, | |
| "eval_runtime": 7.1731, | |
| "eval_samples_per_second": 7.249, | |
| "eval_steps_per_second": 0.976, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02373887240356083, | |
| "grad_norm": 0.2988625315142482, | |
| "learning_rate": 4.705882352941177e-05, | |
| "loss": 1.2538, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02373887240356083, | |
| "eval_loss": 1.117323398590088, | |
| "eval_runtime": 7.1487, | |
| "eval_samples_per_second": 7.274, | |
| "eval_steps_per_second": 0.979, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03560830860534125, | |
| "grad_norm": 0.15105150205016427, | |
| "learning_rate": 7.058823529411765e-05, | |
| "loss": 1.1462, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03560830860534125, | |
| "eval_loss": 1.0231214761734009, | |
| "eval_runtime": 7.1358, | |
| "eval_samples_per_second": 7.287, | |
| "eval_steps_per_second": 0.981, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04747774480712166, | |
| "grad_norm": 0.07095151302909826, | |
| "learning_rate": 9.411764705882353e-05, | |
| "loss": 1.0635, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04747774480712166, | |
| "eval_loss": 0.9480358958244324, | |
| "eval_runtime": 7.1596, | |
| "eval_samples_per_second": 7.263, | |
| "eval_steps_per_second": 0.978, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05934718100890208, | |
| "grad_norm": 0.0510771961177552, | |
| "learning_rate": 0.00011764705882352942, | |
| "loss": 0.9848, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05934718100890208, | |
| "eval_loss": 0.8925089240074158, | |
| "eval_runtime": 7.1518, | |
| "eval_samples_per_second": 7.271, | |
| "eval_steps_per_second": 0.979, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0712166172106825, | |
| "grad_norm": 0.040539214461615646, | |
| "learning_rate": 0.0001411764705882353, | |
| "loss": 0.9513, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0712166172106825, | |
| "eval_loss": 0.8546838164329529, | |
| "eval_runtime": 7.1461, | |
| "eval_samples_per_second": 7.277, | |
| "eval_steps_per_second": 0.98, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0830860534124629, | |
| "grad_norm": 0.038791558909017074, | |
| "learning_rate": 0.0001647058823529412, | |
| "loss": 0.9142, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.0830860534124629, | |
| "eval_loss": 0.8263636231422424, | |
| "eval_runtime": 7.3085, | |
| "eval_samples_per_second": 7.115, | |
| "eval_steps_per_second": 0.958, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.09495548961424333, | |
| "grad_norm": 0.04169274461251293, | |
| "learning_rate": 0.00018823529411764707, | |
| "loss": 0.8854, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09495548961424333, | |
| "eval_loss": 0.8020428419113159, | |
| "eval_runtime": 7.247, | |
| "eval_samples_per_second": 7.175, | |
| "eval_steps_per_second": 0.966, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.10682492581602374, | |
| "grad_norm": 0.043015312738591396, | |
| "learning_rate": 0.00019997847206287532, | |
| "loss": 0.8632, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.10682492581602374, | |
| "eval_loss": 0.7831454873085022, | |
| "eval_runtime": 7.2244, | |
| "eval_samples_per_second": 7.198, | |
| "eval_steps_per_second": 0.969, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.11869436201780416, | |
| "grad_norm": 0.05736189956437236, | |
| "learning_rate": 0.00019980630417613612, | |
| "loss": 0.8669, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11869436201780416, | |
| "eval_loss": 0.7694521546363831, | |
| "eval_runtime": 7.2277, | |
| "eval_samples_per_second": 7.194, | |
| "eval_steps_per_second": 0.968, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13056379821958458, | |
| "grad_norm": 0.04913014792038086, | |
| "learning_rate": 0.0001994622648842964, | |
| "loss": 0.8314, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.13056379821958458, | |
| "eval_loss": 0.7584220170974731, | |
| "eval_runtime": 7.2488, | |
| "eval_samples_per_second": 7.174, | |
| "eval_steps_per_second": 0.966, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.142433234421365, | |
| "grad_norm": 0.04446480334718679, | |
| "learning_rate": 0.00019894694664007728, | |
| "loss": 0.8223, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.142433234421365, | |
| "eval_loss": 0.749190092086792, | |
| "eval_runtime": 7.2447, | |
| "eval_samples_per_second": 7.178, | |
| "eval_steps_per_second": 0.966, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.1543026706231454, | |
| "grad_norm": 0.05216966997945006, | |
| "learning_rate": 0.00019826123684704952, | |
| "loss": 0.8141, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.1543026706231454, | |
| "eval_loss": 0.7411435842514038, | |
| "eval_runtime": 7.2499, | |
| "eval_samples_per_second": 7.173, | |
| "eval_steps_per_second": 0.966, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.1661721068249258, | |
| "grad_norm": 0.04233605582577254, | |
| "learning_rate": 0.00019740631633148045, | |
| "loss": 0.8023, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.1661721068249258, | |
| "eval_loss": 0.7342764139175415, | |
| "eval_runtime": 7.2264, | |
| "eval_samples_per_second": 7.196, | |
| "eval_steps_per_second": 0.969, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.17804154302670624, | |
| "grad_norm": 0.04704056707854844, | |
| "learning_rate": 0.00019638365730889265, | |
| "loss": 0.7895, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.17804154302670624, | |
| "eval_loss": 0.7279341220855713, | |
| "eval_runtime": 7.2397, | |
| "eval_samples_per_second": 7.183, | |
| "eval_steps_per_second": 0.967, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.18991097922848665, | |
| "grad_norm": 0.044026531833322384, | |
| "learning_rate": 0.00019519502084883582, | |
| "loss": 0.7916, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.18991097922848665, | |
| "eval_loss": 0.7229446768760681, | |
| "eval_runtime": 7.2583, | |
| "eval_samples_per_second": 7.164, | |
| "eval_steps_per_second": 0.964, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.20178041543026706, | |
| "grad_norm": 0.05191891103370369, | |
| "learning_rate": 0.00019384245384223767, | |
| "loss": 0.791, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.20178041543026706, | |
| "eval_loss": 0.7180152535438538, | |
| "eval_runtime": 7.2578, | |
| "eval_samples_per_second": 7.165, | |
| "eval_steps_per_second": 0.964, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.21364985163204747, | |
| "grad_norm": 0.05572444654938424, | |
| "learning_rate": 0.00019232828547655615, | |
| "loss": 0.7955, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.21364985163204747, | |
| "eval_loss": 0.7139907479286194, | |
| "eval_runtime": 7.2429, | |
| "eval_samples_per_second": 7.179, | |
| "eval_steps_per_second": 0.966, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.22551928783382788, | |
| "grad_norm": 0.0516837327891167, | |
| "learning_rate": 0.00019065512322480332, | |
| "loss": 0.7776, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.22551928783382788, | |
| "eval_loss": 0.7107158899307251, | |
| "eval_runtime": 7.2537, | |
| "eval_samples_per_second": 7.169, | |
| "eval_steps_per_second": 0.965, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.23738872403560832, | |
| "grad_norm": 0.050846230869752186, | |
| "learning_rate": 0.00018882584835534737, | |
| "loss": 0.785, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.23738872403560832, | |
| "eval_loss": 0.7068641185760498, | |
| "eval_runtime": 7.2578, | |
| "eval_samples_per_second": 7.165, | |
| "eval_steps_per_second": 0.964, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.24925816023738873, | |
| "grad_norm": 0.05343887989356088, | |
| "learning_rate": 0.00018684361097022568, | |
| "loss": 0.7825, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.24925816023738873, | |
| "eval_loss": 0.703826904296875, | |
| "eval_runtime": 7.2465, | |
| "eval_samples_per_second": 7.176, | |
| "eval_steps_per_second": 0.966, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.26112759643916916, | |
| "grad_norm": 0.05242956764694665, | |
| "learning_rate": 0.00018471182458051287, | |
| "loss": 0.7539, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.26112759643916916, | |
| "eval_loss": 0.7010646462440491, | |
| "eval_runtime": 7.2455, | |
| "eval_samples_per_second": 7.177, | |
| "eval_steps_per_second": 0.966, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.27299703264094954, | |
| "grad_norm": 0.051064873049660696, | |
| "learning_rate": 0.00018243416022808547, | |
| "loss": 0.7785, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.27299703264094954, | |
| "eval_loss": 0.6979869604110718, | |
| "eval_runtime": 7.2598, | |
| "eval_samples_per_second": 7.163, | |
| "eval_steps_per_second": 0.964, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.28486646884273, | |
| "grad_norm": 0.046329966712940596, | |
| "learning_rate": 0.00018001454016390586, | |
| "loss": 0.7792, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.28486646884273, | |
| "eval_loss": 0.6954675912857056, | |
| "eval_runtime": 7.2628, | |
| "eval_samples_per_second": 7.16, | |
| "eval_steps_per_second": 0.964, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.29673590504451036, | |
| "grad_norm": 0.05390500282652213, | |
| "learning_rate": 0.00017745713109371139, | |
| "loss": 0.753, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.29673590504451036, | |
| "eval_loss": 0.6930329203605652, | |
| "eval_runtime": 7.2472, | |
| "eval_samples_per_second": 7.175, | |
| "eval_steps_per_second": 0.966, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.3086053412462908, | |
| "grad_norm": 0.043546536548247354, | |
| "learning_rate": 0.00017476633700274068, | |
| "loss": 0.765, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.3086053412462908, | |
| "eval_loss": 0.6907065510749817, | |
| "eval_runtime": 7.2525, | |
| "eval_samples_per_second": 7.17, | |
| "eval_steps_per_second": 0.965, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.32047477744807124, | |
| "grad_norm": 0.06771031218563213, | |
| "learning_rate": 0.00017194679157185255, | |
| "loss": 0.7464, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.32047477744807124, | |
| "eval_loss": 0.6883173584938049, | |
| "eval_runtime": 7.1658, | |
| "eval_samples_per_second": 7.257, | |
| "eval_steps_per_second": 0.977, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.3323442136498516, | |
| "grad_norm": 0.04731592266489432, | |
| "learning_rate": 0.00016900335019809782, | |
| "loss": 0.7649, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.3323442136498516, | |
| "eval_loss": 0.6865535378456116, | |
| "eval_runtime": 7.1591, | |
| "eval_samples_per_second": 7.264, | |
| "eval_steps_per_second": 0.978, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.34421364985163205, | |
| "grad_norm": 0.05422567227576833, | |
| "learning_rate": 0.00016594108163348493, | |
| "loss": 0.764, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.34421364985163205, | |
| "eval_loss": 0.6839883327484131, | |
| "eval_runtime": 7.1803, | |
| "eval_samples_per_second": 7.242, | |
| "eval_steps_per_second": 0.975, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.3560830860534125, | |
| "grad_norm": 0.04716111383242718, | |
| "learning_rate": 0.0001627652592563373, | |
| "loss": 0.7457, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.3560830860534125, | |
| "eval_loss": 0.6814995408058167, | |
| "eval_runtime": 7.1545, | |
| "eval_samples_per_second": 7.268, | |
| "eval_steps_per_second": 0.978, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.36795252225519287, | |
| "grad_norm": 0.04833237070306901, | |
| "learning_rate": 0.00015948135199027474, | |
| "loss": 0.7427, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.36795252225519287, | |
| "eval_loss": 0.6795143485069275, | |
| "eval_runtime": 7.164, | |
| "eval_samples_per_second": 7.258, | |
| "eval_steps_per_second": 0.977, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.3798219584569733, | |
| "grad_norm": 0.04790062682337991, | |
| "learning_rate": 0.00015609501488645554, | |
| "loss": 0.7564, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.3798219584569733, | |
| "eval_loss": 0.6774563193321228, | |
| "eval_runtime": 7.1529, | |
| "eval_samples_per_second": 7.27, | |
| "eval_steps_per_second": 0.979, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.3916913946587537, | |
| "grad_norm": 0.04849257718202548, | |
| "learning_rate": 0.00015261207938529808, | |
| "loss": 0.7352, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.3916913946587537, | |
| "eval_loss": 0.676394522190094, | |
| "eval_runtime": 7.1522, | |
| "eval_samples_per_second": 7.27, | |
| "eval_steps_per_second": 0.979, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.4035608308605341, | |
| "grad_norm": 0.0449523115474861, | |
| "learning_rate": 0.00014903854327445116, | |
| "loss": 0.7471, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.4035608308605341, | |
| "eval_loss": 0.6733577251434326, | |
| "eval_runtime": 7.1516, | |
| "eval_samples_per_second": 7.271, | |
| "eval_steps_per_second": 0.979, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.41543026706231456, | |
| "grad_norm": 0.044399706693004984, | |
| "learning_rate": 0.00014538056036030622, | |
| "loss": 0.7408, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.41543026706231456, | |
| "eval_loss": 0.6722955107688904, | |
| "eval_runtime": 7.1523, | |
| "eval_samples_per_second": 7.27, | |
| "eval_steps_per_second": 0.979, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.42729970326409494, | |
| "grad_norm": 0.04749195230072245, | |
| "learning_rate": 0.00014164442987083762, | |
| "loss": 0.7493, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.42729970326409494, | |
| "eval_loss": 0.6708287000656128, | |
| "eval_runtime": 7.158, | |
| "eval_samples_per_second": 7.265, | |
| "eval_steps_per_second": 0.978, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.4391691394658754, | |
| "grad_norm": 0.04622607097065686, | |
| "learning_rate": 0.0001378365856080198, | |
| "loss": 0.7375, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.4391691394658754, | |
| "eval_loss": 0.6693353056907654, | |
| "eval_runtime": 7.1529, | |
| "eval_samples_per_second": 7.27, | |
| "eval_steps_per_second": 0.979, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.45103857566765576, | |
| "grad_norm": 0.04779103392401475, | |
| "learning_rate": 0.00013396358486850103, | |
| "loss": 0.7469, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.45103857566765576, | |
| "eval_loss": 0.6675475835800171, | |
| "eval_runtime": 7.1609, | |
| "eval_samples_per_second": 7.262, | |
| "eval_steps_per_second": 0.978, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.4629080118694362, | |
| "grad_norm": 0.04367335948535579, | |
| "learning_rate": 0.0001300320971516136, | |
| "loss": 0.7388, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.4629080118694362, | |
| "eval_loss": 0.6652542948722839, | |
| "eval_runtime": 7.1535, | |
| "eval_samples_per_second": 7.269, | |
| "eval_steps_per_second": 0.979, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.47477744807121663, | |
| "grad_norm": 0.04906513653224005, | |
| "learning_rate": 0.0001260488926741651, | |
| "loss": 0.7434, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.47477744807121663, | |
| "eval_loss": 0.6640903949737549, | |
| "eval_runtime": 7.1588, | |
| "eval_samples_per_second": 7.264, | |
| "eval_steps_per_second": 0.978, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.486646884272997, | |
| "grad_norm": 0.04507001139303972, | |
| "learning_rate": 0.00012202083071178938, | |
| "loss": 0.7271, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.486646884272997, | |
| "eval_loss": 0.6632689237594604, | |
| "eval_runtime": 7.1544, | |
| "eval_samples_per_second": 7.268, | |
| "eval_steps_per_second": 0.978, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.49851632047477745, | |
| "grad_norm": 0.054746961897722525, | |
| "learning_rate": 0.00011795484778693382, | |
| "loss": 0.731, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.49851632047477745, | |
| "eval_loss": 0.6615473031997681, | |
| "eval_runtime": 7.1789, | |
| "eval_samples_per_second": 7.243, | |
| "eval_steps_per_second": 0.975, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.5103857566765578, | |
| "grad_norm": 0.04685925852904185, | |
| "learning_rate": 0.00011385794572382357, | |
| "loss": 0.7571, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.5103857566765578, | |
| "eval_loss": 0.660754919052124, | |
| "eval_runtime": 7.1679, | |
| "eval_samples_per_second": 7.255, | |
| "eval_steps_per_second": 0.977, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.5222551928783383, | |
| "grad_norm": 0.05188129116000765, | |
| "learning_rate": 0.00010973717959097327, | |
| "loss": 0.7355, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.5222551928783383, | |
| "eval_loss": 0.6598291993141174, | |
| "eval_runtime": 7.1682, | |
| "eval_samples_per_second": 7.254, | |
| "eval_steps_per_second": 0.977, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.5341246290801187, | |
| "grad_norm": 0.04549793036088958, | |
| "learning_rate": 0.00010559964555200889, | |
| "loss": 0.7379, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5341246290801187, | |
| "eval_loss": 0.658793568611145, | |
| "eval_runtime": 7.1553, | |
| "eval_samples_per_second": 7.267, | |
| "eval_steps_per_second": 0.978, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5459940652818991, | |
| "grad_norm": 0.05162604134236413, | |
| "learning_rate": 0.0001014524686457223, | |
| "loss": 0.7363, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.5459940652818991, | |
| "eval_loss": 0.6571336984634399, | |
| "eval_runtime": 7.156, | |
| "eval_samples_per_second": 7.267, | |
| "eval_steps_per_second": 0.978, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.5578635014836796, | |
| "grad_norm": 0.04783457254038487, | |
| "learning_rate": 9.730279051640112e-05, | |
| "loss": 0.7177, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.5578635014836796, | |
| "eval_loss": 0.655822217464447, | |
| "eval_runtime": 7.1579, | |
| "eval_samples_per_second": 7.265, | |
| "eval_steps_per_second": 0.978, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.56973293768546, | |
| "grad_norm": 0.04436373760940458, | |
| "learning_rate": 9.315775711556298e-05, | |
| "loss": 0.7334, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.56973293768546, | |
| "eval_loss": 0.6540331244468689, | |
| "eval_runtime": 7.1522, | |
| "eval_samples_per_second": 7.271, | |
| "eval_steps_per_second": 0.979, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.5816023738872403, | |
| "grad_norm": 0.04519231140145815, | |
| "learning_rate": 8.902450639627288e-05, | |
| "loss": 0.737, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.5816023738872403, | |
| "eval_loss": 0.6534115672111511, | |
| "eval_runtime": 7.156, | |
| "eval_samples_per_second": 7.267, | |
| "eval_steps_per_second": 0.978, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.5934718100890207, | |
| "grad_norm": 0.04247565082967707, | |
| "learning_rate": 8.491015602123368e-05, | |
| "loss": 0.7413, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5934718100890207, | |
| "eval_loss": 0.6529149413108826, | |
| "eval_runtime": 7.1578, | |
| "eval_samples_per_second": 7.265, | |
| "eval_steps_per_second": 0.978, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6053412462908012, | |
| "grad_norm": 0.048070529053236284, | |
| "learning_rate": 8.082179110581838e-05, | |
| "loss": 0.7141, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.6053412462908012, | |
| "eval_loss": 0.651964545249939, | |
| "eval_runtime": 7.1573, | |
| "eval_samples_per_second": 7.265, | |
| "eval_steps_per_second": 0.978, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.6172106824925816, | |
| "grad_norm": 0.04071752578010737, | |
| "learning_rate": 7.676645201714983e-05, | |
| "loss": 0.7248, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.6172106824925816, | |
| "eval_loss": 0.6511014103889465, | |
| "eval_runtime": 7.1566, | |
| "eval_samples_per_second": 7.266, | |
| "eval_steps_per_second": 0.978, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.629080118694362, | |
| "grad_norm": 0.04168337358828772, | |
| "learning_rate": 7.27511222502395e-05, | |
| "loss": 0.725, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.629080118694362, | |
| "eval_loss": 0.6503429412841797, | |
| "eval_runtime": 7.1538, | |
| "eval_samples_per_second": 7.269, | |
| "eval_steps_per_second": 0.979, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.6409495548961425, | |
| "grad_norm": 0.05260277652628936, | |
| "learning_rate": 6.878271640206281e-05, | |
| "loss": 0.7299, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.6409495548961425, | |
| "eval_loss": 0.6499541401863098, | |
| "eval_runtime": 7.1571, | |
| "eval_samples_per_second": 7.265, | |
| "eval_steps_per_second": 0.978, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.6528189910979229, | |
| "grad_norm": 0.04558705692362712, | |
| "learning_rate": 6.486806826428016e-05, | |
| "loss": 0.7381, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.6528189910979229, | |
| "eval_loss": 0.6490299701690674, | |
| "eval_runtime": 7.1586, | |
| "eval_samples_per_second": 7.264, | |
| "eval_steps_per_second": 0.978, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.6646884272997032, | |
| "grad_norm": 0.046779386203552986, | |
| "learning_rate": 6.101391905510889e-05, | |
| "loss": 0.7052, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.6646884272997032, | |
| "eval_loss": 0.6482182741165161, | |
| "eval_runtime": 7.1587, | |
| "eval_samples_per_second": 7.264, | |
| "eval_steps_per_second": 0.978, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.6765578635014837, | |
| "grad_norm": 0.04433666961158006, | |
| "learning_rate": 5.72269058106111e-05, | |
| "loss": 0.7267, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.6765578635014837, | |
| "eval_loss": 0.6474419832229614, | |
| "eval_runtime": 7.1585, | |
| "eval_samples_per_second": 7.264, | |
| "eval_steps_per_second": 0.978, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.6884272997032641, | |
| "grad_norm": 0.040969171268037584, | |
| "learning_rate": 5.351354995538859e-05, | |
| "loss": 0.7217, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.6884272997032641, | |
| "eval_loss": 0.6467857956886292, | |
| "eval_runtime": 7.1605, | |
| "eval_samples_per_second": 7.262, | |
| "eval_steps_per_second": 0.978, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.7002967359050445, | |
| "grad_norm": 0.04326822029385217, | |
| "learning_rate": 4.988024607236619e-05, | |
| "loss": 0.7225, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.7002967359050445, | |
| "eval_loss": 0.6460662484169006, | |
| "eval_runtime": 7.1561, | |
| "eval_samples_per_second": 7.267, | |
| "eval_steps_per_second": 0.978, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.712166172106825, | |
| "grad_norm": 0.04151648905301352, | |
| "learning_rate": 4.633325089100289e-05, | |
| "loss": 0.7305, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.712166172106825, | |
| "eval_loss": 0.6456657648086548, | |
| "eval_runtime": 7.1693, | |
| "eval_samples_per_second": 7.253, | |
| "eval_steps_per_second": 0.976, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.7240356083086054, | |
| "grad_norm": 0.043211271741518106, | |
| "learning_rate": 4.287867251289348e-05, | |
| "loss": 0.7202, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.7240356083086054, | |
| "eval_loss": 0.6450461745262146, | |
| "eval_runtime": 7.1607, | |
| "eval_samples_per_second": 7.262, | |
| "eval_steps_per_second": 0.978, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.7359050445103857, | |
| "grad_norm": 0.043394130216544674, | |
| "learning_rate": 3.952245989331466e-05, | |
| "loss": 0.7182, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.7359050445103857, | |
| "eval_loss": 0.6443552374839783, | |
| "eval_runtime": 7.1601, | |
| "eval_samples_per_second": 7.262, | |
| "eval_steps_per_second": 0.978, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.7477744807121661, | |
| "grad_norm": 0.0440813623198057, | |
| "learning_rate": 3.627039259682899e-05, | |
| "loss": 0.7269, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.7477744807121661, | |
| "eval_loss": 0.6440042853355408, | |
| "eval_runtime": 7.1528, | |
| "eval_samples_per_second": 7.27, | |
| "eval_steps_per_second": 0.979, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.7596439169139466, | |
| "grad_norm": 0.04196493736245454, | |
| "learning_rate": 3.312807084458831e-05, | |
| "loss": 0.7321, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.7596439169139466, | |
| "eval_loss": 0.6436744332313538, | |
| "eval_runtime": 7.157, | |
| "eval_samples_per_second": 7.266, | |
| "eval_steps_per_second": 0.978, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.771513353115727, | |
| "grad_norm": 0.04409502511550107, | |
| "learning_rate": 3.0100905870475006e-05, | |
| "loss": 0.7127, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.771513353115727, | |
| "eval_loss": 0.6433111429214478, | |
| "eval_runtime": 7.1488, | |
| "eval_samples_per_second": 7.274, | |
| "eval_steps_per_second": 0.979, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.7833827893175074, | |
| "grad_norm": 0.04613176858045456, | |
| "learning_rate": 2.7194110602689026e-05, | |
| "loss": 0.7084, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.7833827893175074, | |
| "eval_loss": 0.6427409648895264, | |
| "eval_runtime": 7.1651, | |
| "eval_samples_per_second": 7.257, | |
| "eval_steps_per_second": 0.977, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.7952522255192879, | |
| "grad_norm": 0.04107118214802092, | |
| "learning_rate": 2.4412690686827e-05, | |
| "loss": 0.7318, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.7952522255192879, | |
| "eval_loss": 0.6424255967140198, | |
| "eval_runtime": 7.151, | |
| "eval_samples_per_second": 7.272, | |
| "eval_steps_per_second": 0.979, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.8071216617210683, | |
| "grad_norm": 0.04076823498258544, | |
| "learning_rate": 2.1761435865912296e-05, | |
| "loss": 0.7111, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.8071216617210683, | |
| "eval_loss": 0.6420266032218933, | |
| "eval_runtime": 7.1562, | |
| "eval_samples_per_second": 7.266, | |
| "eval_steps_per_second": 0.978, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.8189910979228486, | |
| "grad_norm": 0.0402931102905212, | |
| "learning_rate": 1.9244911732219918e-05, | |
| "loss": 0.7173, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.8189910979228486, | |
| "eval_loss": 0.6417113542556763, | |
| "eval_runtime": 7.1559, | |
| "eval_samples_per_second": 7.267, | |
| "eval_steps_per_second": 0.978, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.8308605341246291, | |
| "grad_norm": 0.040268557762468446, | |
| "learning_rate": 1.6867451865100414e-05, | |
| "loss": 0.7187, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.8308605341246291, | |
| "eval_loss": 0.6414252519607544, | |
| "eval_runtime": 7.1614, | |
| "eval_samples_per_second": 7.261, | |
| "eval_steps_per_second": 0.977, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.8427299703264095, | |
| "grad_norm": 0.03910974935669944, | |
| "learning_rate": 1.4633150368341153e-05, | |
| "loss": 0.7003, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.8427299703264095, | |
| "eval_loss": 0.6411222815513611, | |
| "eval_runtime": 7.158, | |
| "eval_samples_per_second": 7.265, | |
| "eval_steps_per_second": 0.978, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.8545994065281899, | |
| "grad_norm": 0.041327420976484716, | |
| "learning_rate": 1.2545854819916646e-05, | |
| "loss": 0.7146, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.8545994065281899, | |
| "eval_loss": 0.6409789323806763, | |
| "eval_runtime": 7.1632, | |
| "eval_samples_per_second": 7.259, | |
| "eval_steps_per_second": 0.977, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.8664688427299704, | |
| "grad_norm": 0.0415664844879509, | |
| "learning_rate": 1.0609159646268506e-05, | |
| "loss": 0.7326, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.8664688427299704, | |
| "eval_loss": 0.640900194644928, | |
| "eval_runtime": 7.1625, | |
| "eval_samples_per_second": 7.26, | |
| "eval_steps_per_second": 0.977, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.8783382789317508, | |
| "grad_norm": 0.040948908512356046, | |
| "learning_rate": 8.82639993252482e-06, | |
| "loss": 0.735, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.8783382789317508, | |
| "eval_loss": 0.6406245827674866, | |
| "eval_runtime": 7.1641, | |
| "eval_samples_per_second": 7.258, | |
| "eval_steps_per_second": 0.977, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.8902077151335311, | |
| "grad_norm": 0.039353242405877606, | |
| "learning_rate": 7.20064567931813e-06, | |
| "loss": 0.7072, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.8902077151335311, | |
| "eval_loss": 0.6405530571937561, | |
| "eval_runtime": 7.1654, | |
| "eval_samples_per_second": 7.257, | |
| "eval_steps_per_second": 0.977, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.9020771513353115, | |
| "grad_norm": 0.040045103519338765, | |
| "learning_rate": 5.734696516092253e-06, | |
| "loss": 0.7242, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.9020771513353115, | |
| "eval_loss": 0.6403719186782837, | |
| "eval_runtime": 7.1641, | |
| "eval_samples_per_second": 7.258, | |
| "eval_steps_per_second": 0.977, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.913946587537092, | |
| "grad_norm": 0.03936062921235335, | |
| "learning_rate": 4.431076880001439e-06, | |
| "loss": 0.725, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.913946587537092, | |
| "eval_loss": 0.6402689218521118, | |
| "eval_runtime": 7.1586, | |
| "eval_samples_per_second": 7.264, | |
| "eval_steps_per_second": 0.978, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.9258160237388724, | |
| "grad_norm": 0.038795606374443216, | |
| "learning_rate": 3.292031668704398e-06, | |
| "loss": 0.7051, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.9258160237388724, | |
| "eval_loss": 0.6402180790901184, | |
| "eval_runtime": 7.1753, | |
| "eval_samples_per_second": 7.247, | |
| "eval_steps_per_second": 0.976, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.9376854599406528, | |
| "grad_norm": 0.04044906534894689, | |
| "learning_rate": 2.3195223745392737e-06, | |
| "loss": 0.7187, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.9376854599406528, | |
| "eval_loss": 0.6401559114456177, | |
| "eval_runtime": 7.178, | |
| "eval_samples_per_second": 7.244, | |
| "eval_steps_per_second": 0.975, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.9495548961424333, | |
| "grad_norm": 0.03933803718026143, | |
| "learning_rate": 1.5152237067365239e-06, | |
| "loss": 0.7185, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.9495548961424333, | |
| "eval_loss": 0.6400689482688904, | |
| "eval_runtime": 7.1614, | |
| "eval_samples_per_second": 7.261, | |
| "eval_steps_per_second": 0.977, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.9614243323442137, | |
| "grad_norm": 0.03894023495076019, | |
| "learning_rate": 8.805207074865873e-07, | |
| "loss": 0.7213, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.9614243323442137, | |
| "eval_loss": 0.6400617957115173, | |
| "eval_runtime": 7.1732, | |
| "eval_samples_per_second": 7.249, | |
| "eval_steps_per_second": 0.976, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.973293768545994, | |
| "grad_norm": 0.03930788498385229, | |
| "learning_rate": 4.165063668285396e-07, | |
| "loss": 0.7192, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.973293768545994, | |
| "eval_loss": 0.6400607228279114, | |
| "eval_runtime": 7.1626, | |
| "eval_samples_per_second": 7.26, | |
| "eval_steps_per_second": 0.977, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.9851632047477745, | |
| "grad_norm": 0.03939532170183159, | |
| "learning_rate": 1.2397974046707283e-07, | |
| "loss": 0.7214, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.9851632047477745, | |
| "eval_loss": 0.6400620341300964, | |
| "eval_runtime": 7.1644, | |
| "eval_samples_per_second": 7.258, | |
| "eval_steps_per_second": 0.977, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.9970326409495549, | |
| "grad_norm": 0.04057041649987638, | |
| "learning_rate": 3.444573758937253e-09, | |
| "loss": 0.7317, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.9970326409495549, | |
| "eval_loss": 0.640075147151947, | |
| "eval_runtime": 7.1623, | |
| "eval_samples_per_second": 7.26, | |
| "eval_steps_per_second": 0.977, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.9994065281899109, | |
| "step": 842, | |
| "total_flos": 6.615025481416704e+16, | |
| "train_loss": 0.6960575034103031, | |
| "train_runtime": 22091.905, | |
| "train_samples_per_second": 2.44, | |
| "train_steps_per_second": 0.038 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 842, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.615025481416704e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |