| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.961832061068702, | |
| "eval_steps": 500, | |
| "global_step": 325, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.015267175572519083, | |
| "grad_norm": 183.11753845214844, | |
| "learning_rate": 6.060606060606061e-06, | |
| "loss": 46.1063, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07633587786259542, | |
| "grad_norm": 136.03738403320312, | |
| "learning_rate": 3.0303030303030306e-05, | |
| "loss": 44.0302, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.15267175572519084, | |
| "grad_norm": 69.2432632446289, | |
| "learning_rate": 6.060606060606061e-05, | |
| "loss": 38.4659, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.22900763358778625, | |
| "grad_norm": 17.486797332763672, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 30.3029, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.3053435114503817, | |
| "grad_norm": 13.530756950378418, | |
| "learning_rate": 0.00012121212121212122, | |
| "loss": 26.6709, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3816793893129771, | |
| "grad_norm": 7.521498680114746, | |
| "learning_rate": 0.00015151515151515152, | |
| "loss": 24.4319, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.4580152671755725, | |
| "grad_norm": 5.912084102630615, | |
| "learning_rate": 0.00018181818181818183, | |
| "loss": 22.862, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5343511450381679, | |
| "grad_norm": 10.610209465026855, | |
| "learning_rate": 0.00019997685019798912, | |
| "loss": 21.5999, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.6106870229007634, | |
| "grad_norm": 20.944725036621094, | |
| "learning_rate": 0.0001997165380022878, | |
| "loss": 19.4719, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6870229007633588, | |
| "grad_norm": 34.12383270263672, | |
| "learning_rate": 0.000199167731989929, | |
| "loss": 14.6832, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.7633587786259542, | |
| "grad_norm": 42.86738204956055, | |
| "learning_rate": 0.0001983320199330545, | |
| "loss": 8.7569, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8396946564885496, | |
| "grad_norm": 12.474686622619629, | |
| "learning_rate": 0.00019721181966290613, | |
| "loss": 4.3457, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.916030534351145, | |
| "grad_norm": 9.623456954956055, | |
| "learning_rate": 0.00019581037207470382, | |
| "loss": 3.4309, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.9923664122137404, | |
| "grad_norm": 3.5216312408447266, | |
| "learning_rate": 0.00019413173175128473, | |
| "loss": 2.9056, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.9923664122137404, | |
| "eval_loss": 2.611328125, | |
| "eval_runtime": 19.2134, | |
| "eval_samples_per_second": 47.935, | |
| "eval_steps_per_second": 0.781, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.0687022900763359, | |
| "grad_norm": 2.9582359790802, | |
| "learning_rate": 0.00019218075523263104, | |
| "loss": 2.7809, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1450381679389312, | |
| "grad_norm": 2.319239616394043, | |
| "learning_rate": 0.00018996308696522433, | |
| "loss": 2.3224, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.2213740458015268, | |
| "grad_norm": 1.3839267492294312, | |
| "learning_rate": 0.00018748514297187648, | |
| "loss": 2.2039, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.297709923664122, | |
| "grad_norm": 0.5840837955474854, | |
| "learning_rate": 0.00018475409228928312, | |
| "loss": 2.1174, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.3740458015267176, | |
| "grad_norm": 1.5493711233139038, | |
| "learning_rate": 0.00018177783622700327, | |
| "loss": 2.0565, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.450381679389313, | |
| "grad_norm": 0.7415986657142639, | |
| "learning_rate": 0.00017856498550787144, | |
| "loss": 2.003, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.5267175572519083, | |
| "grad_norm": 0.6342356204986572, | |
| "learning_rate": 0.00017512483535597867, | |
| "loss": 1.9686, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.6030534351145038, | |
| "grad_norm": 1.0893248319625854, | |
| "learning_rate": 0.00017146733860429612, | |
| "loss": 1.9499, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.6793893129770994, | |
| "grad_norm": 1.233128547668457, | |
| "learning_rate": 0.0001676030768997445, | |
| "loss": 1.9192, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.7557251908396947, | |
| "grad_norm": 0.7829602360725403, | |
| "learning_rate": 0.00016354323008901776, | |
| "loss": 1.8934, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.83206106870229, | |
| "grad_norm": 1.0393383502960205, | |
| "learning_rate": 0.00015929954387373103, | |
| "loss": 1.8579, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.9083969465648853, | |
| "grad_norm": 2.433302879333496, | |
| "learning_rate": 0.00015488429582847192, | |
| "loss": 1.8576, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.984732824427481, | |
| "grad_norm": 1.2537367343902588, | |
| "learning_rate": 0.00015031025988006936, | |
| "loss": 1.8271, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.8229883909225464, | |
| "eval_runtime": 19.0953, | |
| "eval_samples_per_second": 48.232, | |
| "eval_steps_per_second": 0.786, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 2.0610687022900764, | |
| "grad_norm": 1.04417085647583, | |
| "learning_rate": 0.00014559066935084588, | |
| "loss": 1.975, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.1374045801526718, | |
| "grad_norm": 0.9754623174667358, | |
| "learning_rate": 0.00014073917867277557, | |
| "loss": 1.7901, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.213740458015267, | |
| "grad_norm": 0.6031882762908936, | |
| "learning_rate": 0.0001357698238833126, | |
| "loss": 1.7584, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 2.2900763358778624, | |
| "grad_norm": 1.7654844522476196, | |
| "learning_rate": 0.000130696982017182, | |
| "loss": 1.7665, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.366412213740458, | |
| "grad_norm": 1.8184305429458618, | |
| "learning_rate": 0.0001255353295116187, | |
| "loss": 1.7496, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 2.4427480916030535, | |
| "grad_norm": 2.4291305541992188, | |
| "learning_rate": 0.00012029979974539234, | |
| "loss": 1.7389, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.519083969465649, | |
| "grad_norm": 0.7844381928443909, | |
| "learning_rate": 0.00011500553983446527, | |
| "loss": 1.7327, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.595419847328244, | |
| "grad_norm": 1.0221455097198486, | |
| "learning_rate": 0.00010966786680927874, | |
| "loss": 1.7365, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.67175572519084, | |
| "grad_norm": 1.1956524848937988, | |
| "learning_rate": 0.00010430222330045304, | |
| "loss": 1.7204, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.7480916030534353, | |
| "grad_norm": 0.7325518131256104, | |
| "learning_rate": 9.892413286110886e-05, | |
| "loss": 1.7177, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.8244274809160306, | |
| "grad_norm": 0.8538561463356018, | |
| "learning_rate": 9.354915505506839e-05, | |
| "loss": 1.7193, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 2.900763358778626, | |
| "grad_norm": 1.252325415611267, | |
| "learning_rate": 8.81928404408726e-05, | |
| "loss": 1.7058, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.9770992366412212, | |
| "grad_norm": 0.7734937071800232, | |
| "learning_rate": 8.287068558185225e-05, | |
| "loss": 1.7019, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.9923664122137406, | |
| "eval_loss": 1.7041354179382324, | |
| "eval_runtime": 19.3108, | |
| "eval_samples_per_second": 47.694, | |
| "eval_steps_per_second": 0.777, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 3.053435114503817, | |
| "grad_norm": 0.6631619334220886, | |
| "learning_rate": 7.759808821241406e-05, | |
| "loss": 1.8697, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.1297709923664123, | |
| "grad_norm": 0.7187236547470093, | |
| "learning_rate": 7.239030269025311e-05, | |
| "loss": 1.7181, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 3.2061068702290076, | |
| "grad_norm": 0.5320985913276672, | |
| "learning_rate": 6.726239586337408e-05, | |
| "loss": 1.7351, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.282442748091603, | |
| "grad_norm": 0.43638336658477783, | |
| "learning_rate": 6.22292034796035e-05, | |
| "loss": 1.7156, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 3.3587786259541983, | |
| "grad_norm": 0.3966742753982544, | |
| "learning_rate": 5.730528726470792e-05, | |
| "loss": 1.7158, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.435114503816794, | |
| "grad_norm": 0.326159805059433, | |
| "learning_rate": 5.2504892793295e-05, | |
| "loss": 1.7055, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 3.5114503816793894, | |
| "grad_norm": 0.4766685664653778, | |
| "learning_rate": 4.7841908274384616e-05, | |
| "loss": 1.7006, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.5877862595419847, | |
| "grad_norm": 0.41363418102264404, | |
| "learning_rate": 4.332982437088825e-05, | |
| "loss": 1.7106, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 3.66412213740458, | |
| "grad_norm": 0.5006980299949646, | |
| "learning_rate": 3.898169516924398e-05, | |
| "loss": 1.6938, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.7404580152671754, | |
| "grad_norm": 0.4720315933227539, | |
| "learning_rate": 3.4810100412128747e-05, | |
| "loss": 1.6886, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 3.816793893129771, | |
| "grad_norm": 0.5057269334793091, | |
| "learning_rate": 3.0827109103512643e-05, | |
| "loss": 1.6912, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.8931297709923665, | |
| "grad_norm": 0.38378995656967163, | |
| "learning_rate": 2.7044244591351232e-05, | |
| "loss": 1.7001, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 3.969465648854962, | |
| "grad_norm": 0.3008043169975281, | |
| "learning_rate": 2.3472451228937253e-05, | |
| "loss": 1.7024, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.6962379217147827, | |
| "eval_runtime": 18.9852, | |
| "eval_samples_per_second": 48.512, | |
| "eval_steps_per_second": 0.79, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 4.0458015267175576, | |
| "grad_norm": 0.9348434805870056, | |
| "learning_rate": 2.0122062711363532e-05, | |
| "loss": 1.8574, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 4.122137404580153, | |
| "grad_norm": 0.7455368638038635, | |
| "learning_rate": 1.7002772178705716e-05, | |
| "loss": 1.6594, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.198473282442748, | |
| "grad_norm": 0.5774383544921875, | |
| "learning_rate": 1.4123604172419713e-05, | |
| "loss": 1.6527, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 4.2748091603053435, | |
| "grad_norm": 0.5370898842811584, | |
| "learning_rate": 1.149288852608743e-05, | |
| "loss": 1.6587, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.351145038167939, | |
| "grad_norm": 0.7321135997772217, | |
| "learning_rate": 9.118236266049707e-06, | |
| "loss": 1.6676, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 4.427480916030534, | |
| "grad_norm": 0.5155964493751526, | |
| "learning_rate": 7.0065175916482095e-06, | |
| "loss": 1.6579, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.5038167938931295, | |
| "grad_norm": 0.6737932562828064, | |
| "learning_rate": 5.163841998782837e-06, | |
| "loss": 1.6508, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 4.580152671755725, | |
| "grad_norm": 0.9017395377159119, | |
| "learning_rate": 3.595540604290437e-06, | |
| "loss": 1.6375, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.65648854961832, | |
| "grad_norm": 0.5460083484649658, | |
| "learning_rate": 2.30615072228183e-06, | |
| "loss": 1.6522, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 4.732824427480916, | |
| "grad_norm": 0.5443113446235657, | |
| "learning_rate": 1.2994027370611173e-06, | |
| "loss": 1.648, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.809160305343512, | |
| "grad_norm": 0.6177972555160522, | |
| "learning_rate": 5.782093106048159e-07, | |
| "loss": 1.6559, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 4.885496183206107, | |
| "grad_norm": 0.4734289050102234, | |
| "learning_rate": 1.446569558255395e-07, | |
| "loss": 1.6443, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.961832061068702, | |
| "grad_norm": 0.6619871854782104, | |
| "learning_rate": 0.0, | |
| "loss": 1.6463, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 4.961832061068702, | |
| "eval_loss": 1.664337158203125, | |
| "eval_runtime": 18.9808, | |
| "eval_samples_per_second": 48.523, | |
| "eval_steps_per_second": 0.79, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 4.961832061068702, | |
| "step": 325, | |
| "total_flos": 9.909828121379471e+17, | |
| "train_loss": 5.476599056537335, | |
| "train_runtime": 4095.1846, | |
| "train_samples_per_second": 10.222, | |
| "train_steps_per_second": 0.079 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 325, | |
| "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": 9.909828121379471e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |