| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 625, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 16.931909561157227, | |
| "learning_rate": 2.53968253968254e-06, | |
| "loss": 17.7745, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 20.85945701599121, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 16.3129, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 17.975467681884766, | |
| "learning_rate": 8.888888888888888e-06, | |
| "loss": 12.0213, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 6.495634078979492, | |
| "learning_rate": 1.2063492063492064e-05, | |
| "loss": 6.7596, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 5.885069847106934, | |
| "learning_rate": 1.523809523809524e-05, | |
| "loss": 3.8509, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.575489521026611, | |
| "learning_rate": 1.8412698412698415e-05, | |
| "loss": 2.0809, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 5.4581708908081055, | |
| "learning_rate": 1.999609421031453e-05, | |
| "loss": 1.503, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 1.8591735363006592, | |
| "learning_rate": 1.9964866196679105e-05, | |
| "loss": 1.2903, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 2.716611385345459, | |
| "learning_rate": 1.990250772639552e-05, | |
| "loss": 1.174, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.5368252992630005, | |
| "learning_rate": 1.9809213608668188e-05, | |
| "loss": 1.1583, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 1.6447538137435913, | |
| "learning_rate": 1.96852752963305e-05, | |
| "loss": 1.1363, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.368715286254883, | |
| "learning_rate": 1.9531079975339912e-05, | |
| "loss": 1.1077, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 2.465362787246704, | |
| "learning_rate": 1.9347109355200672e-05, | |
| "loss": 1.0916, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 1.520612359046936, | |
| "learning_rate": 1.9133938164092942e-05, | |
| "loss": 1.0992, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 1.3989702463150024, | |
| "learning_rate": 1.8892232353409582e-05, | |
| "loss": 1.0849, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 1.8318055868148804, | |
| "learning_rate": 1.8622747017309676e-05, | |
| "loss": 1.0918, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 1.3566697835922241, | |
| "learning_rate": 1.832632403378808e-05, | |
| "loss": 1.0673, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.595423936843872, | |
| "learning_rate": 1.8003889434630473e-05, | |
| "loss": 1.0789, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 1.4190820455551147, | |
| "learning_rate": 1.765645051247007e-05, | |
| "loss": 1.0618, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 1.8037772178649902, | |
| "learning_rate": 1.728509267398376e-05, | |
| "loss": 1.0658, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 2.374105453491211, | |
| "learning_rate": 1.6890976049058267e-05, | |
| "loss": 1.0481, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 2.712444543838501, | |
| "learning_rate": 1.6475331866519387e-05, | |
| "loss": 1.0856, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 0.846947193145752, | |
| "learning_rate": 1.6039458607746614e-05, | |
| "loss": 1.0575, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 1.0007365942001343, | |
| "learning_rate": 1.558471795018936e-05, | |
| "loss": 1.0391, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.445178747177124, | |
| "learning_rate": 1.5112530513457236e-05, | |
| "loss": 1.0378, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 1.1010501384735107, | |
| "learning_rate": 1.4624371421273823e-05, | |
| "loss": 1.0421, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 1.2421690225601196, | |
| "learning_rate": 1.4121765693158364e-05, | |
| "loss": 1.0279, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 1.585425853729248, | |
| "learning_rate": 1.3606283480231957e-05, | |
| "loss": 1.036, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 1.0381609201431274, | |
| "learning_rate": 1.3079535160031598e-05, | |
| "loss": 1.0348, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.5096782445907593, | |
| "learning_rate": 1.2543166305656099e-05, | |
| "loss": 1.0362, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 0.7411025762557983, | |
| "learning_rate": 1.1998852544960266e-05, | |
| "loss": 1.0293, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 1.0825903415679932, | |
| "learning_rate": 1.1448294325857387e-05, | |
| "loss": 1.0287, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 1.187018871307373, | |
| "learning_rate": 1.0893211604083325e-05, | |
| "loss": 1.0302, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 1.1337244510650635, | |
| "learning_rate": 1.0335338470017742e-05, | |
| "loss": 1.0358, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 1.3198221921920776, | |
| "learning_rate": 9.776417731348416e-06, | |
| "loss": 1.028, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 1.04779851436615, | |
| "learning_rate": 9.218195468502462e-06, | |
| "loss": 1.0221, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 1.228852391242981, | |
| "learning_rate": 8.662415579853492e-06, | |
| "loss": 1.0152, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 0.8538860082626343, | |
| "learning_rate": 8.110814333745496e-06, | |
| "loss": 1.0264, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 1.0245864391326904, | |
| "learning_rate": 7.56511494435318e-06, | |
| "loss": 1.0161, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.983771800994873, | |
| "learning_rate": 7.027022188323716e-06, | |
| "loss": 1.0131, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.2800000000000002, | |
| "grad_norm": 0.8150555491447449, | |
| "learning_rate": 6.498217079017818e-06, | |
| "loss": 1.0271, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 1.072811484336853, | |
| "learning_rate": 5.980351614987759e-06, | |
| "loss": 1.0062, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "grad_norm": 1.0907390117645264, | |
| "learning_rate": 5.475043619098334e-06, | |
| "loss": 1.0137, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "grad_norm": 0.8887478113174438, | |
| "learning_rate": 4.983871684413363e-06, | |
| "loss": 0.9897, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.8817629218101501, | |
| "learning_rate": 4.508370242636968e-06, | |
| "loss": 1.0112, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "grad_norm": 1.7927525043487549, | |
| "learning_rate": 4.050024770515869e-06, | |
| "loss": 0.9977, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "grad_norm": 1.2345198392868042, | |
| "learning_rate": 3.6102671491780393e-06, | |
| "loss": 0.9906, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.9548463225364685, | |
| "learning_rate": 3.1904711909051933e-06, | |
| "loss": 1.0106, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "grad_norm": 1.2012273073196411, | |
| "learning_rate": 2.7919483473136678e-06, | |
| "loss": 0.9938, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.8137465715408325, | |
| "learning_rate": 2.4159436123512737e-06, | |
| "loss": 0.9991, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 0.9726725220680237, | |
| "learning_rate": 2.0636316329094317e-06, | |
| "loss": 0.9852, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "grad_norm": 0.9943102598190308, | |
| "learning_rate": 1.7361130392009407e-06, | |
| "loss": 0.9887, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "grad_norm": 0.8023040890693665, | |
| "learning_rate": 1.4344110063675143e-06, | |
| "loss": 0.9813, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "grad_norm": 0.987560510635376, | |
| "learning_rate": 1.1594680580585815e-06, | |
| "loss": 0.9934, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 0.8008418083190918, | |
| "learning_rate": 9.121431219671096e-07, | |
| "loss": 0.9944, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "grad_norm": 1.4193716049194336, | |
| "learning_rate": 6.932088465209941e-07, | |
| "loss": 0.9935, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.5600000000000005, | |
| "grad_norm": 1.3066496849060059, | |
| "learning_rate": 5.033491871127105e-07, | |
| "loss": 0.9825, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "grad_norm": 0.9865421056747437, | |
| "learning_rate": 3.4315726940795436e-07, | |
| "loss": 0.9849, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "grad_norm": 0.7975912690162659, | |
| "learning_rate": 2.1313353640827207e-07, | |
| "loss": 0.9803, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.8560538291931152, | |
| "learning_rate": 1.1368418505635303e-07, | |
| "loss": 0.9821, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "grad_norm": 0.9261242747306824, | |
| "learning_rate": 4.5119897268023347e-08, | |
| "loss": 0.9878, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "grad_norm": 0.7569891214370728, | |
| "learning_rate": 7.654869355252503e-09, | |
| "loss": 0.9881, | |
| "step": 620 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 625, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.425513594532659e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |