| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.6021505376344085, | |
| "eval_steps": 1000, | |
| "global_step": 7200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0011947431302270011, | |
| "grad_norm": 23.375, | |
| "learning_rate": 2e-06, | |
| "loss": 2.4074, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.11947431302270012, | |
| "grad_norm": 0.515625, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7151, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.23894862604540024, | |
| "grad_norm": 0.369140625, | |
| "learning_rate": 0.0004, | |
| "loss": 0.4333, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.35842293906810035, | |
| "grad_norm": 0.361328125, | |
| "learning_rate": 0.0006, | |
| "loss": 0.4113, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4778972520908005, | |
| "grad_norm": 0.244140625, | |
| "learning_rate": 0.0008, | |
| "loss": 0.393, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5973715651135006, | |
| "grad_norm": 0.2099609375, | |
| "learning_rate": 0.001, | |
| "loss": 0.3857, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7168458781362007, | |
| "grad_norm": 0.193359375, | |
| "learning_rate": 0.0012, | |
| "loss": 0.3734, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.8363201911589009, | |
| "grad_norm": 0.1796875, | |
| "learning_rate": 0.0014, | |
| "loss": 0.373, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.955794504181601, | |
| "grad_norm": 0.126953125, | |
| "learning_rate": 0.0016, | |
| "loss": 0.3578, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.075268817204301, | |
| "grad_norm": 0.111328125, | |
| "learning_rate": 0.0018000000000000002, | |
| "loss": 0.3473, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.1947431302270013, | |
| "grad_norm": 0.125, | |
| "learning_rate": 0.002, | |
| "loss": 0.3551, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.3142174432497014, | |
| "grad_norm": 0.1591796875, | |
| "learning_rate": 0.0019987165071710528, | |
| "loss": 0.3466, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.4336917562724014, | |
| "grad_norm": 0.099609375, | |
| "learning_rate": 0.001994869323391895, | |
| "loss": 0.3251, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.5531660692951015, | |
| "grad_norm": 0.10888671875, | |
| "learning_rate": 0.0019884683243281114, | |
| "loss": 0.3183, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.6726403823178018, | |
| "grad_norm": 0.1337890625, | |
| "learning_rate": 0.0019795299412524946, | |
| "loss": 0.3006, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.7921146953405018, | |
| "grad_norm": 0.1494140625, | |
| "learning_rate": 0.001968077118866204, | |
| "loss": 0.2911, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.911589008363202, | |
| "grad_norm": 0.1201171875, | |
| "learning_rate": 0.001954139256400049, | |
| "loss": 0.2853, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.031063321385902, | |
| "grad_norm": 0.08154296875, | |
| "learning_rate": 0.0019377521321470807, | |
| "loss": 0.2585, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.150537634408602, | |
| "grad_norm": 0.12890625, | |
| "learning_rate": 0.0019189578116202307, | |
| "loss": 0.2611, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.270011947431302, | |
| "grad_norm": 0.125, | |
| "learning_rate": 0.0018978045395707417, | |
| "loss": 0.2502, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.3894862604540026, | |
| "grad_norm": 0.1162109375, | |
| "learning_rate": 0.0018743466161445822, | |
| "loss": 0.2455, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.5089605734767026, | |
| "grad_norm": 0.1181640625, | |
| "learning_rate": 0.001848644257494751, | |
| "loss": 0.2292, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.6284348864994027, | |
| "grad_norm": 0.1318359375, | |
| "learning_rate": 0.0018207634412072764, | |
| "loss": 0.2199, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.7479091995221028, | |
| "grad_norm": 0.12060546875, | |
| "learning_rate": 0.0017907757369376985, | |
| "loss": 0.203, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.867383512544803, | |
| "grad_norm": 0.1064453125, | |
| "learning_rate": 0.0017587581226927908, | |
| "loss": 0.2035, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.986857825567503, | |
| "grad_norm": 0.09228515625, | |
| "learning_rate": 0.00172479278722912, | |
| "loss": 0.1938, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.106332138590203, | |
| "grad_norm": 0.095703125, | |
| "learning_rate": 0.0016889669190756867, | |
| "loss": 0.1868, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.225806451612903, | |
| "grad_norm": 0.07666015625, | |
| "learning_rate": 0.0016513724827222226, | |
| "loss": 0.1884, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.3452807646356035, | |
| "grad_norm": 0.08837890625, | |
| "learning_rate": 0.001612105982547663, | |
| "loss": 0.1772, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.4647550776583036, | |
| "grad_norm": 0.10693359375, | |
| "learning_rate": 0.0015712682150947923, | |
| "loss": 0.1694, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.5842293906810037, | |
| "grad_norm": 0.0927734375, | |
| "learning_rate": 0.0015289640103269624, | |
| "loss": 0.1616, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.7037037037037037, | |
| "grad_norm": 0.06201171875, | |
| "learning_rate": 0.0014853019625310814, | |
| "loss": 0.1519, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 3.823178016726404, | |
| "grad_norm": 0.2275390625, | |
| "learning_rate": 0.0014403941515576345, | |
| "loss": 0.1488, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.942652329749104, | |
| "grad_norm": 0.09765625, | |
| "learning_rate": 0.0013943558551133186, | |
| "loss": 0.1455, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.062126642771804, | |
| "grad_norm": 0.080078125, | |
| "learning_rate": 0.0013473052528448202, | |
| "loss": 0.1354, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.181600955794504, | |
| "grad_norm": 0.08349609375, | |
| "learning_rate": 0.0012993631229733583, | |
| "loss": 0.1386, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.301075268817204, | |
| "grad_norm": 0.095703125, | |
| "learning_rate": 0.0012506525322587205, | |
| "loss": 0.1355, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.4205495818399045, | |
| "grad_norm": 0.06982421875, | |
| "learning_rate": 0.0012012985200886602, | |
| "loss": 0.1246, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 4.540023894862604, | |
| "grad_norm": 0.07958984375, | |
| "learning_rate": 0.0011514277775045768, | |
| "loss": 0.1213, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 4.659498207885305, | |
| "grad_norm": 0.0810546875, | |
| "learning_rate": 0.0011011683219874322, | |
| "loss": 0.1157, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 4.778972520908005, | |
| "grad_norm": 0.07763671875, | |
| "learning_rate": 0.0010506491688387128, | |
| "loss": 0.1106, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.898446833930705, | |
| "grad_norm": 0.06591796875, | |
| "learning_rate": 0.001, | |
| "loss": 0.109, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 5.017921146953405, | |
| "grad_norm": 0.08935546875, | |
| "learning_rate": 0.0009493508311612874, | |
| "loss": 0.1051, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 5.137395459976105, | |
| "grad_norm": 0.0771484375, | |
| "learning_rate": 0.000898831678012568, | |
| "loss": 0.1028, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 5.256869772998805, | |
| "grad_norm": 0.08056640625, | |
| "learning_rate": 0.0008485722224954237, | |
| "loss": 0.1024, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 5.376344086021505, | |
| "grad_norm": 0.08837890625, | |
| "learning_rate": 0.0007987014799113398, | |
| "loss": 0.0981, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.4958183990442055, | |
| "grad_norm": 0.09228515625, | |
| "learning_rate": 0.0007493474677412793, | |
| "loss": 0.0937, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 5.615292712066906, | |
| "grad_norm": 0.0712890625, | |
| "learning_rate": 0.000700636877026642, | |
| "loss": 0.0908, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 5.734767025089606, | |
| "grad_norm": 0.06298828125, | |
| "learning_rate": 0.0006526947471551798, | |
| "loss": 0.0865, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 5.854241338112306, | |
| "grad_norm": 0.06396484375, | |
| "learning_rate": 0.0006056441448866817, | |
| "loss": 0.0867, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 5.973715651135006, | |
| "grad_norm": 0.056884765625, | |
| "learning_rate": 0.0005596058484423656, | |
| "loss": 0.0849, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.093189964157706, | |
| "grad_norm": 0.06640625, | |
| "learning_rate": 0.0005146980374689192, | |
| "loss": 0.0812, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 6.212664277180406, | |
| "grad_norm": 0.08984375, | |
| "learning_rate": 0.0004710359896730378, | |
| "loss": 0.0825, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 6.332138590203106, | |
| "grad_norm": 0.057373046875, | |
| "learning_rate": 0.00042873178490520746, | |
| "loss": 0.0818, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 6.451612903225806, | |
| "grad_norm": 0.057861328125, | |
| "learning_rate": 0.0003878940174523371, | |
| "loss": 0.0769, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 6.571087216248507, | |
| "grad_norm": 0.05517578125, | |
| "learning_rate": 0.00034862751727777797, | |
| "loss": 0.0768, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 6.690561529271207, | |
| "grad_norm": 0.056396484375, | |
| "learning_rate": 0.00031103308092431337, | |
| "loss": 0.0749, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 6.810035842293907, | |
| "grad_norm": 0.044189453125, | |
| "learning_rate": 0.00027520721277088026, | |
| "loss": 0.0746, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 6.929510155316607, | |
| "grad_norm": 0.041015625, | |
| "learning_rate": 0.00024124187730720914, | |
| "loss": 0.0748, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 7.048984468339307, | |
| "grad_norm": 0.047607421875, | |
| "learning_rate": 0.00020922426306230158, | |
| "loss": 0.0736, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 7.168458781362007, | |
| "grad_norm": 0.047607421875, | |
| "learning_rate": 0.00017923655879272394, | |
| "loss": 0.0743, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.287933094384707, | |
| "grad_norm": 0.06787109375, | |
| "learning_rate": 0.00015135574250524898, | |
| "loss": 0.076, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 7.407407407407407, | |
| "grad_norm": 0.06201171875, | |
| "learning_rate": 0.00012565338385541792, | |
| "loss": 0.0736, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 7.526881720430108, | |
| "grad_norm": 0.047607421875, | |
| "learning_rate": 0.00010219546042925842, | |
| "loss": 0.0728, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 7.646356033452808, | |
| "grad_norm": 0.045654296875, | |
| "learning_rate": 8.10421883797694e-05, | |
| "loss": 0.0725, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 7.765830346475508, | |
| "grad_norm": 0.05126953125, | |
| "learning_rate": 6.22478678529197e-05, | |
| "loss": 0.0717, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 7.885304659498208, | |
| "grad_norm": 0.046875, | |
| "learning_rate": 4.5860743599951184e-05, | |
| "loss": 0.0726, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 1.004778972520908, | |
| "grad_norm": 0.050048828125, | |
| "learning_rate": 3.192288113379582e-05, | |
| "loss": 0.0732, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 1.124253285543608, | |
| "grad_norm": 0.0498046875, | |
| "learning_rate": 2.0470058747505516e-05, | |
| "loss": 0.0724, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 1.2437275985663083, | |
| "grad_norm": 0.050537109375, | |
| "learning_rate": 1.153167567188862e-05, | |
| "loss": 0.0748, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 1.3632019115890084, | |
| "grad_norm": 0.04443359375, | |
| "learning_rate": 5.1306766081048454e-06, | |
| "loss": 0.0745, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.4826762246117084, | |
| "grad_norm": 0.050048828125, | |
| "learning_rate": 1.2834928289472415e-06, | |
| "loss": 0.0717, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 1.6021505376344085, | |
| "grad_norm": 0.0458984375, | |
| "learning_rate": 0.0, | |
| "loss": 0.073, | |
| "step": 7200 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 7200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 2000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.018481533644288e+17, | |
| "train_batch_size": 12, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |