| { | |
| "best_metric": 0.6282580494880676, | |
| "best_model_checkpoint": "./qlora-out/checkpoint-21000", | |
| "epoch": 0.782968569404571, | |
| "global_step": 21000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 0.00019999938245325715, | |
| "loss": 0.9023, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00019999724773356797, | |
| "loss": 0.8027, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.0001999935882494411, | |
| "loss": 0.8041, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 0.00019998840405667672, | |
| "loss": 0.7944, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.00019998169523432365, | |
| "loss": 0.81, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 0.0001999734618846785, | |
| "loss": 0.7855, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.00019996370413328385, | |
| "loss": 0.7849, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.00019995242212892653, | |
| "loss": 0.7564, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.00019993961604363532, | |
| "loss": 0.7724, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00019992528607267815, | |
| "loss": 0.7308, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_loss": 0.7677998542785645, | |
| "eval_runtime": 1774.3517, | |
| "eval_samples_per_second": 0.305, | |
| "eval_steps_per_second": 0.305, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0001999094324345594, | |
| "loss": 0.7844, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.00019989205537101633, | |
| "loss": 0.7668, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00019987315514701553, | |
| "loss": 0.7727, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00019985273205074878, | |
| "loss": 0.7467, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00019983078639362883, | |
| "loss": 0.7516, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00019980731851028445, | |
| "loss": 0.7267, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0001997823287585554, | |
| "loss": 0.7632, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.000199755817519487, | |
| "loss": 0.7392, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00019972778519732436, | |
| "loss": 0.7528, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0001996982322195061, | |
| "loss": 0.725, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 0.7452704310417175, | |
| "eval_runtime": 1787.7554, | |
| "eval_samples_per_second": 0.303, | |
| "eval_steps_per_second": 0.303, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00019966715903665795, | |
| "loss": 0.7234, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00019963456612258576, | |
| "loss": 0.754, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00019960045397426841, | |
| "loss": 0.7856, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00019956482311185006, | |
| "loss": 0.7387, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00019952767407863245, | |
| "loss": 0.7309, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00019948900744106633, | |
| "loss": 0.7232, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00019944882378874316, | |
| "loss": 0.7406, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.0001994071237343858, | |
| "loss": 0.7166, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00019936390791383936, | |
| "loss": 0.7308, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00019931917698606143, | |
| "loss": 0.7288, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_loss": 0.7343490123748779, | |
| "eval_runtime": 1770.9966, | |
| "eval_samples_per_second": 0.306, | |
| "eval_steps_per_second": 0.306, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00019927293163311206, | |
| "loss": 0.7236, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00019922517256014337, | |
| "loss": 0.716, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00019917590049538874, | |
| "loss": 0.7564, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00019912511619015177, | |
| "loss": 0.7082, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00019907282041879484, | |
| "loss": 0.7103, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00019901901397872715, | |
| "loss": 0.7457, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0001989636976903928, | |
| "loss": 0.7076, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0001989068723972581, | |
| "loss": 0.7217, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00019884853896579873, | |
| "loss": 0.7175, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0001987886982854866, | |
| "loss": 0.7083, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 0.726176917552948, | |
| "eval_runtime": 1765.3933, | |
| "eval_samples_per_second": 0.307, | |
| "eval_steps_per_second": 0.307, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00019872735126877622, | |
| "loss": 0.7228, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0001986644988510909, | |
| "loss": 0.7133, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00019860014199080822, | |
| "loss": 0.7243, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00019853428166924576, | |
| "loss": 0.6929, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.00019846691889064593, | |
| "loss": 0.7392, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0001983980546821607, | |
| "loss": 0.7247, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0001983276900938359, | |
| "loss": 0.7258, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00019825582619859532, | |
| "loss": 0.7197, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0001981824640922242, | |
| "loss": 0.6906, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00019810760489335266, | |
| "loss": 0.7274, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 0.7171670794487, | |
| "eval_runtime": 1812.7597, | |
| "eval_samples_per_second": 0.299, | |
| "eval_steps_per_second": 0.299, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0001980312497434385, | |
| "loss": 0.7105, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00019795339980675002, | |
| "loss": 0.7091, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.00019787405627034804, | |
| "loss": 0.7102, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0001977932203440678, | |
| "loss": 0.7314, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00019771089326050075, | |
| "loss": 0.6945, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.0001976270762749755, | |
| "loss": 0.7048, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00019754177066553882, | |
| "loss": 0.6963, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00019745497773293613, | |
| "loss": 0.711, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0001973666988005916, | |
| "loss": 0.7017, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00019727693521458806, | |
| "loss": 0.7287, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 0.710155725479126, | |
| "eval_runtime": 1786.7467, | |
| "eval_samples_per_second": 0.303, | |
| "eval_steps_per_second": 0.303, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00019718568834364638, | |
| "loss": 0.6894, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00019709295957910476, | |
| "loss": 0.7061, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.00019699875033489728, | |
| "loss": 0.7063, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.00019690306204753254, | |
| "loss": 0.6872, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0001968058961760717, | |
| "loss": 0.7095, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00019670725420210618, | |
| "loss": 0.695, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0001966071376297351, | |
| "loss": 0.674, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00019650554798554236, | |
| "loss": 0.7225, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00019640248681857342, | |
| "loss": 0.6845, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00019629795570031149, | |
| "loss": 0.6891, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_loss": 0.703677773475647, | |
| "eval_runtime": 1767.3593, | |
| "eval_samples_per_second": 0.307, | |
| "eval_steps_per_second": 0.307, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00019619195622465379, | |
| "loss": 0.6962, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0001960844900078871, | |
| "loss": 0.6779, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00019597555868866318, | |
| "loss": 0.7354, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00019586516392797374, | |
| "loss": 0.7196, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0001957533074091252, | |
| "loss": 0.682, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.0001956399908377129, | |
| "loss": 0.6938, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0001955252159415952, | |
| "loss": 0.6912, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00019540898447086705, | |
| "loss": 0.7048, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.00019529129819783334, | |
| "loss": 0.7007, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00019517215891698192, | |
| "loss": 0.6969, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.6973471641540527, | |
| "eval_runtime": 1793.2355, | |
| "eval_samples_per_second": 0.302, | |
| "eval_steps_per_second": 0.302, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00019505156844495619, | |
| "loss": 0.6894, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00019492952862052733, | |
| "loss": 0.6971, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0001948060413045665, | |
| "loss": 0.7135, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0001946811083800161, | |
| "loss": 0.6794, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0001945547317518614, | |
| "loss": 0.7086, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00019442691334710136, | |
| "loss": 0.7042, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00019429765511471916, | |
| "loss": 0.6822, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.0001941669590256526, | |
| "loss": 0.7016, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00019403482707276406, | |
| "loss": 0.705, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00019390126127080999, | |
| "loss": 0.698, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_loss": 0.6910382509231567, | |
| "eval_runtime": 1782.1661, | |
| "eval_samples_per_second": 0.304, | |
| "eval_steps_per_second": 0.304, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00019376626365641026, | |
| "loss": 0.6926, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.0001936298362880172, | |
| "loss": 0.6871, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00019349198124588403, | |
| "loss": 0.6894, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00019335270063203325, | |
| "loss": 0.6894, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00019321199657022464, | |
| "loss": 0.7057, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00019306987120592265, | |
| "loss": 0.6682, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00019292632670626401, | |
| "loss": 0.6931, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00019278136526002443, | |
| "loss": 0.7244, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.0001926349890775853, | |
| "loss": 0.6881, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00019248720039090006, | |
| "loss": 0.6839, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": 0.6857322454452515, | |
| "eval_runtime": 1760.8664, | |
| "eval_samples_per_second": 0.308, | |
| "eval_steps_per_second": 0.308, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00019233800145346006, | |
| "loss": 0.6917, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0001921873945402602, | |
| "loss": 0.6672, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00019203538194776442, | |
| "loss": 0.6873, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00019188196599387043, | |
| "loss": 0.6733, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00019172714901787453, | |
| "loss": 0.706, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00019157093338043583, | |
| "loss": 0.6848, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00019141332146354042, | |
| "loss": 0.6728, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.00019125431567046494, | |
| "loss": 0.686, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00019109391842573987, | |
| "loss": 0.6992, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.00019093213217511265, | |
| "loss": 0.6675, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_loss": 0.6794907450675964, | |
| "eval_runtime": 1782.5413, | |
| "eval_samples_per_second": 0.304, | |
| "eval_steps_per_second": 0.304, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0001907689593855104, | |
| "loss": 0.6721, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00019060440254500228, | |
| "loss": 0.6353, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.00019043846416276155, | |
| "loss": 0.6449, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.0001902711467690272, | |
| "loss": 0.6451, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00019010245291506569, | |
| "loss": 0.6421, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00018993238517313167, | |
| "loss": 0.6352, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0001897609461364289, | |
| "loss": 0.6371, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00018958813841907083, | |
| "loss": 0.623, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00018941396465604063, | |
| "loss": 0.6533, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00018923842750315095, | |
| "loss": 0.6371, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_loss": 0.6759930849075317, | |
| "eval_runtime": 1304.1351, | |
| "eval_samples_per_second": 0.416, | |
| "eval_steps_per_second": 0.416, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00018906152963700358, | |
| "loss": 0.6664, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00018888327375494847, | |
| "loss": 0.6644, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00018870366257504274, | |
| "loss": 0.623, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00018852269883600918, | |
| "loss": 0.6242, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00018834038529719446, | |
| "loss": 0.6318, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0001881567247385271, | |
| "loss": 0.6359, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00018797171996047505, | |
| "loss": 0.6449, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00018778537378400304, | |
| "loss": 0.6434, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00018759768905052946, | |
| "loss": 0.6323, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00018740866862188317, | |
| "loss": 0.6377, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 0.6696639060974121, | |
| "eval_runtime": 1238.6847, | |
| "eval_samples_per_second": 0.438, | |
| "eval_steps_per_second": 0.438, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0001872183153802598, | |
| "loss": 0.6232, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00018702663222817774, | |
| "loss": 0.6236, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00018683362208843395, | |
| "loss": 0.6331, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00018663928790405945, | |
| "loss": 0.6528, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.00018644363263827426, | |
| "loss": 0.6362, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00018624665927444248, | |
| "loss": 0.6308, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00018604837081602656, | |
| "loss": 0.6107, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00018584877028654154, | |
| "loss": 0.6418, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00018564786072950917, | |
| "loss": 0.6276, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00018544564520841118, | |
| "loss": 0.6296, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_loss": 0.6651941537857056, | |
| "eval_runtime": 1277.2835, | |
| "eval_samples_per_second": 0.424, | |
| "eval_steps_per_second": 0.424, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00018524212680664286, | |
| "loss": 0.636, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00018503730862746574, | |
| "loss": 0.6643, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00018483119379396058, | |
| "loss": 0.6282, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0001846237854489796, | |
| "loss": 0.6381, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00018441508675509844, | |
| "loss": 0.6692, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00018420510089456823, | |
| "loss": 0.6478, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00018399383106926676, | |
| "loss": 0.6293, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00018378128050064988, | |
| "loss": 0.6406, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0001835674524297023, | |
| "loss": 0.6407, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.0001833523501168881, | |
| "loss": 0.633, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_loss": 0.6601429581642151, | |
| "eval_runtime": 1275.6071, | |
| "eval_samples_per_second": 0.425, | |
| "eval_steps_per_second": 0.425, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00018313597684210115, | |
| "loss": 0.6198, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00018291833590461498, | |
| "loss": 0.6345, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00018269943062303257, | |
| "loss": 0.6554, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.00018247926433523562, | |
| "loss": 0.6151, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00018225784039833386, | |
| "loss": 0.6331, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0001820351621886136, | |
| "loss": 0.6256, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.0001818112331014865, | |
| "loss": 0.6263, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00018158605655143757, | |
| "loss": 0.6015, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00018135963597197327, | |
| "loss": 0.6144, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00018113197481556912, | |
| "loss": 0.613, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.6547831892967224, | |
| "eval_runtime": 1305.9645, | |
| "eval_samples_per_second": 0.415, | |
| "eval_steps_per_second": 0.415, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00018090307655361701, | |
| "loss": 0.6354, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.00018067294467637228, | |
| "loss": 0.6349, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00018044158269290054, | |
| "loss": 0.6127, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00018020899413102412, | |
| "loss": 0.5977, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00017997518253726834, | |
| "loss": 0.6213, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00017974015147680734, | |
| "loss": 0.6168, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00017950390453340978, | |
| "loss": 0.5978, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.0001792664453093842, | |
| "loss": 0.6201, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.000179027777425524, | |
| "loss": 0.6141, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00017878790452105245, | |
| "loss": 0.6135, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 0.6480616927146912, | |
| "eval_runtime": 1347.9883, | |
| "eval_samples_per_second": 0.402, | |
| "eval_steps_per_second": 0.402, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0001785468302535669, | |
| "loss": 0.6363, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00017830455829898317, | |
| "loss": 0.6076, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00017806109235147963, | |
| "loss": 0.609, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00017781643612344058, | |
| "loss": 0.6044, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00017757059334539994, | |
| "loss": 0.6262, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.00017732356776598403, | |
| "loss": 0.6195, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0001770753631518548, | |
| "loss": 0.6328, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.000176825983287652, | |
| "loss": 0.6028, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0001765754319759358, | |
| "loss": 0.6159, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0001763237130371287, | |
| "loss": 0.6169, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 0.6444052457809448, | |
| "eval_runtime": 1304.3701, | |
| "eval_samples_per_second": 0.416, | |
| "eval_steps_per_second": 0.416, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0001760708303094572, | |
| "loss": 0.6183, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00017581678764889324, | |
| "loss": 0.6116, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.00017556158892909567, | |
| "loss": 0.6406, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00017530523804135085, | |
| "loss": 0.6223, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00017504773889451361, | |
| "loss": 0.628, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00017478909541494736, | |
| "loss": 0.6173, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00017452931154646444, | |
| "loss": 0.61, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00017426839125026598, | |
| "loss": 0.5959, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00017400633850488128, | |
| "loss": 0.5979, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00017374315730610745, | |
| "loss": 0.6161, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 0.6378119587898254, | |
| "eval_runtime": 1283.5987, | |
| "eval_samples_per_second": 0.422, | |
| "eval_steps_per_second": 0.422, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00017347885166694825, | |
| "loss": 0.6213, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00017321342561755297, | |
| "loss": 0.6217, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00017294688320515506, | |
| "loss": 0.6127, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00017267922849401024, | |
| "loss": 0.6145, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00017241046556533472, | |
| "loss": 0.5936, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0001721405985172428, | |
| "loss": 0.6273, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0001718696314646846, | |
| "loss": 0.6059, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.000171597568539383, | |
| "loss": 0.5934, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.000171324413889771, | |
| "loss": 0.6243, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00017105017168092808, | |
| "loss": 0.6164, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.6324757933616638, | |
| "eval_runtime": 1266.6769, | |
| "eval_samples_per_second": 0.428, | |
| "eval_steps_per_second": 0.428, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.0001707748460945171, | |
| "loss": 0.5953, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.0001704984413287202, | |
| "loss": 0.6329, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00017022096159817493, | |
| "loss": 0.6227, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00016994241113391003, | |
| "loss": 0.6022, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.0001696627941832808, | |
| "loss": 0.604, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0001693821150099044, | |
| "loss": 0.6101, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00016910037789359485, | |
| "loss": 0.6242, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00016881758713029776, | |
| "loss": 0.6096, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0001685337470320248, | |
| "loss": 0.5948, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0001682488619267879, | |
| "loss": 0.5911, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_loss": 0.6282580494880676, | |
| "eval_runtime": 1313.1215, | |
| "eval_samples_per_second": 0.413, | |
| "eval_steps_per_second": 0.413, | |
| "step": 21000 | |
| } | |
| ], | |
| "max_steps": 80463, | |
| "num_train_epochs": 3, | |
| "total_flos": 5.888261973449687e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |