| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9987628865979383, | |
| "eval_steps": 500, | |
| "global_step": 909, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.032989690721649485, | |
| "grad_norm": 3.190283465974621, | |
| "learning_rate": 5e-06, | |
| "loss": 1.0361, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06597938144329897, | |
| "grad_norm": 1.2412136578455273, | |
| "learning_rate": 5e-06, | |
| "loss": 0.9152, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09896907216494845, | |
| "grad_norm": 0.9937983329085553, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8771, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13195876288659794, | |
| "grad_norm": 0.9565722646796444, | |
| "learning_rate": 5e-06, | |
| "loss": 0.855, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.16494845360824742, | |
| "grad_norm": 0.8422015410274669, | |
| "learning_rate": 5e-06, | |
| "loss": 0.834, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1979381443298969, | |
| "grad_norm": 0.9927761059685116, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8249, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2309278350515464, | |
| "grad_norm": 1.0124625468461579, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8092, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2639175257731959, | |
| "grad_norm": 0.9162707322543721, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8002, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.29690721649484536, | |
| "grad_norm": 0.7438936667406614, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7953, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.32989690721649484, | |
| "grad_norm": 0.74943987165883, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7912, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3628865979381443, | |
| "grad_norm": 0.6393536784375358, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7835, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3958762886597938, | |
| "grad_norm": 0.6732184370236527, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7846, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.4288659793814433, | |
| "grad_norm": 0.675715599863121, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7835, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4618556701030928, | |
| "grad_norm": 0.638472159519646, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7797, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4948453608247423, | |
| "grad_norm": 1.1212007213917934, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7759, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5278350515463918, | |
| "grad_norm": 0.5929058650136099, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7758, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5608247422680412, | |
| "grad_norm": 0.7325664578284696, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7746, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5938144329896907, | |
| "grad_norm": 0.7146407893100764, | |
| "learning_rate": 5e-06, | |
| "loss": 0.772, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.6268041237113402, | |
| "grad_norm": 0.6297828498939105, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7686, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6597938144329897, | |
| "grad_norm": 0.650337063259678, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7668, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6927835051546392, | |
| "grad_norm": 0.577352278155154, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7633, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.7257731958762886, | |
| "grad_norm": 0.6351053699389445, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7606, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.7587628865979381, | |
| "grad_norm": 0.6179445706530043, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7644, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7917525773195876, | |
| "grad_norm": 0.7772047208925177, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7585, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.8247422680412371, | |
| "grad_norm": 0.6393400921262609, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7597, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.8577319587628865, | |
| "grad_norm": 0.5366628092052804, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7559, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8907216494845361, | |
| "grad_norm": 0.6897451596502111, | |
| "learning_rate": 5e-06, | |
| "loss": 0.757, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.9237113402061856, | |
| "grad_norm": 0.683076031456689, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7595, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.9567010309278351, | |
| "grad_norm": 0.6342586759859082, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7576, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.9896907216494846, | |
| "grad_norm": 0.6339977397184522, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7548, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9995876288659794, | |
| "eval_loss": 0.7519278526306152, | |
| "eval_runtime": 322.6674, | |
| "eval_samples_per_second": 25.314, | |
| "eval_steps_per_second": 0.397, | |
| "step": 303 | |
| }, | |
| { | |
| "epoch": 1.022680412371134, | |
| "grad_norm": 0.810990287818241, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7934, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.0556701030927835, | |
| "grad_norm": 0.5989713675220099, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7044, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.088659793814433, | |
| "grad_norm": 0.584782182855064, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7115, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.1216494845360825, | |
| "grad_norm": 0.7858920415247334, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7094, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.1546391752577319, | |
| "grad_norm": 0.9401995798606461, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7079, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.1876288659793814, | |
| "grad_norm": 0.6150009311102699, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7073, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.220618556701031, | |
| "grad_norm": 0.6009149100944755, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7096, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.2536082474226804, | |
| "grad_norm": 0.6115518108906659, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7066, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.2865979381443298, | |
| "grad_norm": 0.7496882281145417, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7076, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.3195876288659794, | |
| "grad_norm": 0.6685224897984725, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7062, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.352577319587629, | |
| "grad_norm": 0.641185927057492, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7117, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.3855670103092783, | |
| "grad_norm": 0.5361388827305237, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7094, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.418556701030928, | |
| "grad_norm": 1.002359631516242, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7054, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.4515463917525773, | |
| "grad_norm": 0.8431450479727091, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7075, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.4845360824742269, | |
| "grad_norm": 0.6447323729739957, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7099, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.5175257731958762, | |
| "grad_norm": 0.8431314429320579, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7018, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.5505154639175258, | |
| "grad_norm": 0.6273662519128372, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7051, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.5835051546391754, | |
| "grad_norm": 0.8396735090007554, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7106, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.6164948453608248, | |
| "grad_norm": 0.5802654475284174, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7043, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.6494845360824741, | |
| "grad_norm": 0.6224806599884348, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7086, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.6824742268041237, | |
| "grad_norm": 0.6154446076130442, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7026, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.7154639175257733, | |
| "grad_norm": 0.5857753449684375, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7037, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.7484536082474227, | |
| "grad_norm": 0.5716099691987403, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7066, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.781443298969072, | |
| "grad_norm": 0.6774790897099987, | |
| "learning_rate": 5e-06, | |
| "loss": 0.707, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.8144329896907216, | |
| "grad_norm": 0.6117062221128381, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6996, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.8474226804123712, | |
| "grad_norm": 0.5362825787566358, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7036, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.8804123711340206, | |
| "grad_norm": 0.7851595485494056, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7066, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.91340206185567, | |
| "grad_norm": 0.642752104749906, | |
| "learning_rate": 5e-06, | |
| "loss": 0.705, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.9463917525773196, | |
| "grad_norm": 0.6663997639727156, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7051, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.9793814432989691, | |
| "grad_norm": 0.6435653630361237, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7026, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.9991752577319588, | |
| "eval_loss": 0.7376570701599121, | |
| "eval_runtime": 321.676, | |
| "eval_samples_per_second": 25.392, | |
| "eval_steps_per_second": 0.398, | |
| "step": 606 | |
| }, | |
| { | |
| "epoch": 2.0123711340206185, | |
| "grad_norm": 0.7966992322635892, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7418, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.045360824742268, | |
| "grad_norm": 0.5919842716689093, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6581, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.0783505154639177, | |
| "grad_norm": 0.6225831303900108, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6494, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.111340206185567, | |
| "grad_norm": 0.5720666970317613, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6557, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.1443298969072164, | |
| "grad_norm": 0.625621284764116, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6562, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.177319587628866, | |
| "grad_norm": 0.722621026378947, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6592, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.2103092783505156, | |
| "grad_norm": 0.6611874958125228, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6576, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.243298969072165, | |
| "grad_norm": 0.624720046082098, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6534, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.2762886597938143, | |
| "grad_norm": 0.6227890769590231, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6561, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.3092783505154637, | |
| "grad_norm": 0.6353543358518403, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6564, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.3422680412371135, | |
| "grad_norm": 0.616682251013517, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6558, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.375257731958763, | |
| "grad_norm": 0.5785627398529801, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6579, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.4082474226804123, | |
| "grad_norm": 0.7087632640527876, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6578, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.441237113402062, | |
| "grad_norm": 0.7221097669514308, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6555, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.4742268041237114, | |
| "grad_norm": 0.6845092133296887, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6589, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.507216494845361, | |
| "grad_norm": 0.6131735355128494, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6597, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.54020618556701, | |
| "grad_norm": 0.74499117668607, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6604, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.5731958762886595, | |
| "grad_norm": 0.6953072761863929, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6599, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.6061855670103093, | |
| "grad_norm": 0.7683634702318719, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6584, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.6391752577319587, | |
| "grad_norm": 0.9202931242949187, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6599, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.6721649484536085, | |
| "grad_norm": 0.6785534766587453, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6605, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.705154639175258, | |
| "grad_norm": 0.9373759072613878, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6633, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.7381443298969073, | |
| "grad_norm": 0.5316447851690145, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6582, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.7711340206185566, | |
| "grad_norm": 0.7810499110998566, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6638, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.804123711340206, | |
| "grad_norm": 0.5581690358208933, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6572, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.837113402061856, | |
| "grad_norm": 0.5757480690524878, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6629, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.870103092783505, | |
| "grad_norm": 0.6570394054126519, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6603, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.9030927835051545, | |
| "grad_norm": 0.5532161107989387, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6596, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.9360824742268044, | |
| "grad_norm": 0.6779485831959426, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6615, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.9690721649484537, | |
| "grad_norm": 0.6105580266011457, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6573, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.9987628865979383, | |
| "eval_loss": 0.737443208694458, | |
| "eval_runtime": 321.3627, | |
| "eval_samples_per_second": 25.417, | |
| "eval_steps_per_second": 0.398, | |
| "step": 909 | |
| }, | |
| { | |
| "epoch": 2.9987628865979383, | |
| "step": 909, | |
| "total_flos": 1522399476449280.0, | |
| "train_loss": 0.7216839113644641, | |
| "train_runtime": 53747.0627, | |
| "train_samples_per_second": 8.662, | |
| "train_steps_per_second": 0.017 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 909, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1522399476449280.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |