| { | |
| "best_metric": 0.6857322454452515, | |
| "best_model_checkpoint": "./qlora-out/checkpoint-10000", | |
| "epoch": 0.3728421759069386, | |
| "global_step": 10000, | |
| "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 | |
| } | |
| ], | |
| "max_steps": 80463, | |
| "num_train_epochs": 3, | |
| "total_flos": 2.803380622630748e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |