| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 100, | |
| "global_step": 630, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.047619047619047616, | |
| "grad_norm": 0.6868933439254761, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.2872, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": 0.7597203850746155, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.2256, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.14285714285714285, | |
| "grad_norm": 0.5019340515136719, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.2407, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 0.7778149247169495, | |
| "learning_rate": 6.349206349206349e-05, | |
| "loss": 0.24, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.23809523809523808, | |
| "grad_norm": 0.5550121665000916, | |
| "learning_rate": 7.936507936507937e-05, | |
| "loss": 0.253, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 0.5827834606170654, | |
| "learning_rate": 9.523809523809524e-05, | |
| "loss": 0.244, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 0.3167480230331421, | |
| "learning_rate": 9.996239762521151e-05, | |
| "loss": 0.201, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 0.4390418827533722, | |
| "learning_rate": 9.977835886801848e-05, | |
| "loss": 0.215, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.42857142857142855, | |
| "grad_norm": 0.4620651304721832, | |
| "learning_rate": 9.944154131125642e-05, | |
| "loss": 0.2332, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "grad_norm": 0.6008896231651306, | |
| "learning_rate": 9.895297870901379e-05, | |
| "loss": 0.2197, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.47619047619047616, | |
| "eval_question_finetune_val_loss": 0.22019629180431366, | |
| "eval_question_finetune_val_runtime": 147.8155, | |
| "eval_question_finetune_val_samples_per_second": 1.15, | |
| "eval_question_finetune_val_steps_per_second": 1.15, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5238095238095238, | |
| "grad_norm": 0.41202816367149353, | |
| "learning_rate": 9.83141705484087e-05, | |
| "loss": 0.2104, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 0.5663377046585083, | |
| "learning_rate": 9.752707744739145e-05, | |
| "loss": 0.23, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.6190476190476191, | |
| "grad_norm": 0.3894096910953522, | |
| "learning_rate": 9.659411513725284e-05, | |
| "loss": 0.2026, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 0.5098657608032227, | |
| "learning_rate": 9.551814704830734e-05, | |
| "loss": 0.2058, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 0.6123200058937073, | |
| "learning_rate": 9.430247552150673e-05, | |
| "loss": 0.2148, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.7619047619047619, | |
| "grad_norm": 0.3542161285877228, | |
| "learning_rate": 9.295083167295766e-05, | |
| "loss": 0.1923, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.8095238095238095, | |
| "grad_norm": 0.5978957414627075, | |
| "learning_rate": 9.146736394245047e-05, | |
| "loss": 0.2044, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 0.5118722915649414, | |
| "learning_rate": 8.985662536114613e-05, | |
| "loss": 0.1983, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.9047619047619048, | |
| "grad_norm": 0.3764272630214691, | |
| "learning_rate": 8.812355957749882e-05, | |
| "loss": 0.1601, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "grad_norm": 0.3428772985935211, | |
| "learning_rate": 8.627348568430352e-05, | |
| "loss": 0.21, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9523809523809523, | |
| "eval_question_finetune_val_loss": 0.2040266990661621, | |
| "eval_question_finetune_val_runtime": 147.8548, | |
| "eval_question_finetune_val_samples_per_second": 1.15, | |
| "eval_question_finetune_val_steps_per_second": 1.15, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.3707384169101715, | |
| "learning_rate": 8.43120818934367e-05, | |
| "loss": 0.191, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.0476190476190477, | |
| "grad_norm": 0.32814162969589233, | |
| "learning_rate": 8.224536810839564e-05, | |
| "loss": 0.1219, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.0952380952380953, | |
| "grad_norm": 0.3846438527107239, | |
| "learning_rate": 8.007968744812424e-05, | |
| "loss": 0.1382, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 0.600942075252533, | |
| "learning_rate": 7.782168677883206e-05, | |
| "loss": 0.1216, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.1904761904761905, | |
| "grad_norm": 0.3708088994026184, | |
| "learning_rate": 7.547829631355794e-05, | |
| "loss": 0.1294, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.2380952380952381, | |
| "grad_norm": 0.45602136850357056, | |
| "learning_rate": 7.3056708342091e-05, | |
| "loss": 0.1235, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.2857142857142856, | |
| "grad_norm": 0.4373123347759247, | |
| "learning_rate": 7.056435515653059e-05, | |
| "loss": 0.1144, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.3915819227695465, | |
| "learning_rate": 6.800888624023553e-05, | |
| "loss": 0.1182, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.380952380952381, | |
| "grad_norm": 0.3638611435890198, | |
| "learning_rate": 6.539814479017428e-05, | |
| "loss": 0.1161, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 0.4916077256202698, | |
| "learning_rate": 6.274014364473274e-05, | |
| "loss": 0.1284, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "eval_question_finetune_val_loss": 0.20424266159534454, | |
| "eval_question_finetune_val_runtime": 147.8685, | |
| "eval_question_finetune_val_samples_per_second": 1.15, | |
| "eval_question_finetune_val_steps_per_second": 1.15, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.4761904761904763, | |
| "grad_norm": 0.32184702157974243, | |
| "learning_rate": 6.004304069086196e-05, | |
| "loss": 0.1174, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.5238095238095237, | |
| "grad_norm": 0.3934004306793213, | |
| "learning_rate": 5.731511382604544e-05, | |
| "loss": 0.1203, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.5714285714285714, | |
| "grad_norm": 0.32937076687812805, | |
| "learning_rate": 5.456473555193242e-05, | |
| "loss": 0.1267, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.619047619047619, | |
| "grad_norm": 0.44530630111694336, | |
| "learning_rate": 5.180034727761382e-05, | |
| "loss": 0.1295, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.5469300746917725, | |
| "learning_rate": 4.903043341140879e-05, | |
| "loss": 0.1275, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.42091014981269836, | |
| "learning_rate": 4.626349532067879e-05, | |
| "loss": 0.131, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.7619047619047619, | |
| "grad_norm": 0.5508981943130493, | |
| "learning_rate": 4.350802523959138e-05, | |
| "loss": 0.1259, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.8095238095238095, | |
| "grad_norm": 0.536870539188385, | |
| "learning_rate": 4.0772480204915364e-05, | |
| "loss": 0.1361, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.8571428571428572, | |
| "grad_norm": 0.39201387763023376, | |
| "learning_rate": 3.806525609984312e-05, | |
| "loss": 0.0992, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "grad_norm": 0.4120243191719055, | |
| "learning_rate": 3.5394661885504386e-05, | |
| "loss": 0.1388, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.9047619047619047, | |
| "eval_question_finetune_val_loss": 0.19780363142490387, | |
| "eval_question_finetune_val_runtime": 147.8694, | |
| "eval_question_finetune_val_samples_per_second": 1.15, | |
| "eval_question_finetune_val_steps_per_second": 1.15, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.9523809523809523, | |
| "grad_norm": 0.6459566354751587, | |
| "learning_rate": 3.276889409925963e-05, | |
| "loss": 0.1227, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.5514885187149048, | |
| "learning_rate": 3.019601169804216e-05, | |
| "loss": 0.124, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.0476190476190474, | |
| "grad_norm": 0.29513123631477356, | |
| "learning_rate": 2.7683911323959712e-05, | |
| "loss": 0.0804, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.0952380952380953, | |
| "grad_norm": 0.4728158116340637, | |
| "learning_rate": 2.524030306806944e-05, | |
| "loss": 0.0834, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "grad_norm": 0.35146838426589966, | |
| "learning_rate": 2.2872686806712035e-05, | |
| "loss": 0.0664, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.1904761904761907, | |
| "grad_norm": 0.3647899627685547, | |
| "learning_rate": 2.0588329183032744e-05, | |
| "loss": 0.0637, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.238095238095238, | |
| "grad_norm": 0.3702828884124756, | |
| "learning_rate": 1.8394241304337413e-05, | |
| "loss": 0.0694, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.27304157614707947, | |
| "learning_rate": 1.629715722373423e-05, | |
| "loss": 0.0586, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.3333333333333335, | |
| "grad_norm": 0.3929966688156128, | |
| "learning_rate": 1.4303513272105057e-05, | |
| "loss": 0.0653, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "grad_norm": 0.3267512619495392, | |
| "learning_rate": 1.2419428303840091e-05, | |
| "loss": 0.0639, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.380952380952381, | |
| "eval_question_finetune_val_loss": 0.2098911553621292, | |
| "eval_question_finetune_val_runtime": 147.8662, | |
| "eval_question_finetune_val_samples_per_second": 1.15, | |
| "eval_question_finetune_val_steps_per_second": 1.15, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.4285714285714284, | |
| "grad_norm": 0.3743651211261749, | |
| "learning_rate": 1.0650684916965559e-05, | |
| "loss": 0.0655, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.4761904761904763, | |
| "grad_norm": 0.6065438389778137, | |
| "learning_rate": 9.002711705302886e-06, | |
| "loss": 0.0664, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.5238095238095237, | |
| "grad_norm": 0.38518133759498596, | |
| "learning_rate": 7.48056659713105e-06, | |
| "loss": 0.0618, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.2961275279521942, | |
| "learning_rate": 6.088921331488568e-06, | |
| "loss": 0.0614, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.619047619047619, | |
| "grad_norm": 0.3589339554309845, | |
| "learning_rate": 4.832047119760186e-06, | |
| "loss": 0.0579, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 0.4137968122959137, | |
| "learning_rate": 3.7138015365554833e-06, | |
| "loss": 0.0657, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.7142857142857144, | |
| "grad_norm": 0.5474274158477783, | |
| "learning_rate": 2.737616680113758e-06, | |
| "loss": 0.0655, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.761904761904762, | |
| "grad_norm": 0.6094250679016113, | |
| "learning_rate": 1.9064886385728399e-06, | |
| "loss": 0.0646, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.8095238095238093, | |
| "grad_norm": 0.3034960925579071, | |
| "learning_rate": 1.2229682944319031e-06, | |
| "loss": 0.0618, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.4673740863800049, | |
| "learning_rate": 6.891534954310885e-07, | |
| "loss": 0.0709, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_question_finetune_val_loss": 0.21026459336280823, | |
| "eval_question_finetune_val_runtime": 147.9364, | |
| "eval_question_finetune_val_samples_per_second": 1.149, | |
| "eval_question_finetune_val_steps_per_second": 1.149, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.9047619047619047, | |
| "grad_norm": 0.43377596139907837, | |
| "learning_rate": 3.066826158767855e-07, | |
| "loss": 0.0658, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.9523809523809526, | |
| "grad_norm": 0.4616926312446594, | |
| "learning_rate": 7.672952817413537e-08, | |
| "loss": 0.0632, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.3746221959590912, | |
| "learning_rate": 0.0, | |
| "loss": 0.0606, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 630, | |
| "total_flos": 2.11652989949952e+16, | |
| "train_loss": 0.1360657389201815, | |
| "train_runtime": 7264.9105, | |
| "train_samples_per_second": 0.347, | |
| "train_steps_per_second": 0.087 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 630, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 400, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.11652989949952e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |