| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.968354430379748, | |
| "eval_steps": 500, | |
| "global_step": 1575, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15822784810126583, | |
| "grad_norm": 2.5220770835876465, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3375, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.31645569620253167, | |
| "grad_norm": 1.2550034523010254, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4176, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.47468354430379744, | |
| "grad_norm": 1.1539708375930786, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0198, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.6329113924050633, | |
| "grad_norm": 1.0697557926177979, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2321, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7911392405063291, | |
| "grad_norm": 1.0829753875732422, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9474, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.9493670886075949, | |
| "grad_norm": 1.0121601819992065, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8395, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1075949367088607, | |
| "grad_norm": 1.1056463718414307, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7928, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.2658227848101267, | |
| "grad_norm": 1.5227317810058594, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4643, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.4240506329113924, | |
| "grad_norm": 0.5029327273368835, | |
| "learning_rate": 0.0002, | |
| "loss": 0.574, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.5822784810126582, | |
| "grad_norm": 0.7477303147315979, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5327, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.740506329113924, | |
| "grad_norm": 1.1760517358779907, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3683, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.8987341772151898, | |
| "grad_norm": 0.7232487797737122, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6297, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.0569620253164556, | |
| "grad_norm": 0.8394080996513367, | |
| "learning_rate": 0.0002, | |
| "loss": 0.445, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.2151898734177213, | |
| "grad_norm": 0.7312848567962646, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3746, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.3734177215189876, | |
| "grad_norm": 0.7731461524963379, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4716, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.5316455696202533, | |
| "grad_norm": 1.2274295091629028, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3655, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.689873417721519, | |
| "grad_norm": 0.6810910105705261, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3526, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.848101265822785, | |
| "grad_norm": 0.6517865657806396, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4433, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.0063291139240507, | |
| "grad_norm": 1.4362159967422485, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3225, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 3.1645569620253164, | |
| "grad_norm": 0.5174972414970398, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4203, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.3227848101265822, | |
| "grad_norm": 0.8872693181037903, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3485, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 3.481012658227848, | |
| "grad_norm": 0.7814120054244995, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2683, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.6392405063291138, | |
| "grad_norm": 0.5629352331161499, | |
| "learning_rate": 0.0002, | |
| "loss": 0.4093, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 3.7974683544303796, | |
| "grad_norm": 0.8676897883415222, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3342, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.9556962025316453, | |
| "grad_norm": 0.3980434536933899, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2799, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 4.113924050632911, | |
| "grad_norm": 0.6955259442329407, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3424, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.272151898734177, | |
| "grad_norm": 0.9159439206123352, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2344, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 4.430379746835443, | |
| "grad_norm": 0.657145082950592, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3157, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.588607594936709, | |
| "grad_norm": 0.6370977163314819, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3165, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 4.746835443037975, | |
| "grad_norm": 1.4098331928253174, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2306, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.905063291139241, | |
| "grad_norm": 0.49123522639274597, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3595, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 5.063291139240507, | |
| "grad_norm": 0.6442649960517883, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2739, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.2215189873417724, | |
| "grad_norm": 0.3584900498390198, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2148, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 5.379746835443038, | |
| "grad_norm": 0.6529756784439087, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3141, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.537974683544304, | |
| "grad_norm": 0.7614108324050903, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2484, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 5.69620253164557, | |
| "grad_norm": 0.37845131754875183, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2489, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.8544303797468356, | |
| "grad_norm": 0.5563950538635254, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3095, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 6.012658227848101, | |
| "grad_norm": 0.9632635712623596, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2197, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.170886075949367, | |
| "grad_norm": 0.32439491152763367, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2652, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 6.329113924050633, | |
| "grad_norm": 0.5222536325454712, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2422, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.487341772151899, | |
| "grad_norm": 0.5735016465187073, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1888, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 6.6455696202531644, | |
| "grad_norm": 0.6607617735862732, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2995, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 6.80379746835443, | |
| "grad_norm": 0.9632560610771179, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2397, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 6.962025316455696, | |
| "grad_norm": 0.25032752752304077, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2035, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.120253164556962, | |
| "grad_norm": 0.4147493243217468, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2388, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 7.2784810126582276, | |
| "grad_norm": 0.6790456771850586, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1916, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 7.436708860759493, | |
| "grad_norm": 0.29760053753852844, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2266, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 7.594936708860759, | |
| "grad_norm": 0.6803601980209351, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2337, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 7.753164556962025, | |
| "grad_norm": 0.8090017437934875, | |
| "learning_rate": 0.0002, | |
| "loss": 0.199, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 7.911392405063291, | |
| "grad_norm": 0.4255053699016571, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2577, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.069620253164556, | |
| "grad_norm": 0.6785822510719299, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2045, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 8.227848101265822, | |
| "grad_norm": 0.2853826880455017, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1707, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.386075949367088, | |
| "grad_norm": 0.5013365745544434, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2324, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 8.544303797468354, | |
| "grad_norm": 0.7551653385162354, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2055, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 8.70253164556962, | |
| "grad_norm": 0.2432267665863037, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2046, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 8.860759493670885, | |
| "grad_norm": 0.6924687027931213, | |
| "learning_rate": 0.0002, | |
| "loss": 0.2369, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.018987341772151, | |
| "grad_norm": 0.6701030135154724, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1816, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 9.177215189873417, | |
| "grad_norm": 0.3679279088973999, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1884, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 9.335443037974684, | |
| "grad_norm": 0.3734745979309082, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1934, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 9.49367088607595, | |
| "grad_norm": 0.3465384244918823, | |
| "learning_rate": 0.0002, | |
| "loss": 0.167, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 9.651898734177216, | |
| "grad_norm": 0.4366382360458374, | |
| "learning_rate": 0.0002, | |
| "loss": 0.244, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 9.810126582278482, | |
| "grad_norm": 0.44247567653656006, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1965, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 9.968354430379748, | |
| "grad_norm": 0.2957931458950043, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1794, | |
| "step": 1575 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1580, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 25, | |
| "total_flos": 1.479668134699008e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |