| { |
| "best_metric": 0.57263505, |
| "best_model_checkpoint": "/export/home2/zli/kc/mm_rag/Qwen2.5-14B-Instruct_lora/checkpoint-256", |
| "epoch": 0.9978075517661389, |
| "eval_steps": 100, |
| "global_step": 256, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.00389768574908648, |
| "grad_norm": 0.3756321966648102, |
| "learning_rate": 7.692307692307694e-06, |
| "loss": 0.760678231716156, |
| "memory(GiB)": 64.79, |
| "step": 1, |
| "token_acc": 0.794377180381695, |
| "train_speed(iter/s)": 0.008942 |
| }, |
| { |
| "epoch": 0.0194884287454324, |
| "grad_norm": 0.4006075859069824, |
| "learning_rate": 3.846153846153846e-05, |
| "loss": 0.8511521816253662, |
| "memory(GiB)": 69.21, |
| "step": 5, |
| "token_acc": 0.7628462851792697, |
| "train_speed(iter/s)": 0.009262 |
| }, |
| { |
| "epoch": 0.0389768574908648, |
| "grad_norm": 0.40541699528694153, |
| "learning_rate": 7.692307692307693e-05, |
| "loss": 0.8799286842346191, |
| "memory(GiB)": 69.21, |
| "step": 10, |
| "token_acc": 0.755640420079393, |
| "train_speed(iter/s)": 0.009459 |
| }, |
| { |
| "epoch": 0.058465286236297195, |
| "grad_norm": 0.28694042563438416, |
| "learning_rate": 9.998328666948438e-05, |
| "loss": 0.7268682479858398, |
| "memory(GiB)": 69.21, |
| "step": 15, |
| "token_acc": 0.7857048253940223, |
| "train_speed(iter/s)": 0.009584 |
| }, |
| { |
| "epoch": 0.0779537149817296, |
| "grad_norm": 0.29239076375961304, |
| "learning_rate": 9.979538999730047e-05, |
| "loss": 0.6990458011627197, |
| "memory(GiB)": 69.21, |
| "step": 20, |
| "token_acc": 0.7923689929968607, |
| "train_speed(iter/s)": 0.009633 |
| }, |
| { |
| "epoch": 0.09744214372716199, |
| "grad_norm": 0.25992509722709656, |
| "learning_rate": 9.939949247384046e-05, |
| "loss": 0.7058870792388916, |
| "memory(GiB)": 71.47, |
| "step": 25, |
| "token_acc": 0.7879081525173794, |
| "train_speed(iter/s)": 0.009549 |
| }, |
| { |
| "epoch": 0.11693057247259439, |
| "grad_norm": 0.1924816370010376, |
| "learning_rate": 9.879724780684519e-05, |
| "loss": 0.7370085716247559, |
| "memory(GiB)": 71.47, |
| "step": 30, |
| "token_acc": 0.7818104986383698, |
| "train_speed(iter/s)": 0.009632 |
| }, |
| { |
| "epoch": 0.1364190012180268, |
| "grad_norm": 0.2375946342945099, |
| "learning_rate": 9.799117163889559e-05, |
| "loss": 0.6951110363006592, |
| "memory(GiB)": 71.47, |
| "step": 35, |
| "token_acc": 0.7915360501567398, |
| "train_speed(iter/s)": 0.009685 |
| }, |
| { |
| "epoch": 0.1559074299634592, |
| "grad_norm": 0.19938509166240692, |
| "learning_rate": 9.698463103929542e-05, |
| "loss": 0.634189510345459, |
| "memory(GiB)": 71.47, |
| "step": 40, |
| "token_acc": 0.8057028291378926, |
| "train_speed(iter/s)": 0.009727 |
| }, |
| { |
| "epoch": 0.1753958587088916, |
| "grad_norm": 0.2671877145767212, |
| "learning_rate": 9.57818304394503e-05, |
| "loss": 0.6699491500854492, |
| "memory(GiB)": 71.47, |
| "step": 45, |
| "token_acc": 0.8010264675941635, |
| "train_speed(iter/s)": 0.009776 |
| }, |
| { |
| "epoch": 0.19488428745432398, |
| "grad_norm": 0.3463566303253174, |
| "learning_rate": 9.438779407049281e-05, |
| "loss": 0.6760294437408447, |
| "memory(GiB)": 71.47, |
| "step": 50, |
| "token_acc": 0.793778801843318, |
| "train_speed(iter/s)": 0.009808 |
| }, |
| { |
| "epoch": 0.2143727161997564, |
| "grad_norm": 0.22926518321037292, |
| "learning_rate": 9.280834497651334e-05, |
| "loss": 0.6952155113220215, |
| "memory(GiB)": 71.47, |
| "step": 55, |
| "token_acc": 0.7856186209763081, |
| "train_speed(iter/s)": 0.009819 |
| }, |
| { |
| "epoch": 0.23386114494518878, |
| "grad_norm": 0.26200923323631287, |
| "learning_rate": 9.105008069106093e-05, |
| "loss": 0.6922572135925293, |
| "memory(GiB)": 71.47, |
| "step": 60, |
| "token_acc": 0.784200469035004, |
| "train_speed(iter/s)": 0.009839 |
| }, |
| { |
| "epoch": 0.25334957369062117, |
| "grad_norm": 0.2608520984649658, |
| "learning_rate": 8.912034567851599e-05, |
| "loss": 0.6162422180175782, |
| "memory(GiB)": 71.47, |
| "step": 65, |
| "token_acc": 0.8122029543994862, |
| "train_speed(iter/s)": 0.009864 |
| }, |
| { |
| "epoch": 0.2728380024360536, |
| "grad_norm": 0.23958414793014526, |
| "learning_rate": 8.702720065545024e-05, |
| "loss": 0.6201987266540527, |
| "memory(GiB)": 71.47, |
| "step": 70, |
| "token_acc": 0.807689317368994, |
| "train_speed(iter/s)": 0.00982 |
| }, |
| { |
| "epoch": 0.292326431181486, |
| "grad_norm": 0.24707700312137604, |
| "learning_rate": 8.47793889201221e-05, |
| "loss": 0.6751769542694092, |
| "memory(GiB)": 71.47, |
| "step": 75, |
| "token_acc": 0.7933965994464215, |
| "train_speed(iter/s)": 0.009843 |
| }, |
| { |
| "epoch": 0.3118148599269184, |
| "grad_norm": 0.280536413192749, |
| "learning_rate": 8.238629983075294e-05, |
| "loss": 0.6537846088409424, |
| "memory(GiB)": 71.47, |
| "step": 80, |
| "token_acc": 0.800564772326156, |
| "train_speed(iter/s)": 0.00983 |
| }, |
| { |
| "epoch": 0.3313032886723508, |
| "grad_norm": 0.29046493768692017, |
| "learning_rate": 7.985792958513931e-05, |
| "loss": 0.6765205383300781, |
| "memory(GiB)": 71.47, |
| "step": 85, |
| "token_acc": 0.7913814331445457, |
| "train_speed(iter/s)": 0.009828 |
| }, |
| { |
| "epoch": 0.3507917174177832, |
| "grad_norm": 0.2879982888698578, |
| "learning_rate": 7.720483946542914e-05, |
| "loss": 0.6516193866729736, |
| "memory(GiB)": 71.47, |
| "step": 90, |
| "token_acc": 0.7983723394009439, |
| "train_speed(iter/s)": 0.009817 |
| }, |
| { |
| "epoch": 0.3702801461632156, |
| "grad_norm": 0.28098541498184204, |
| "learning_rate": 7.443811172247821e-05, |
| "loss": 0.6455563545227051, |
| "memory(GiB)": 71.47, |
| "step": 95, |
| "token_acc": 0.8004983012457532, |
| "train_speed(iter/s)": 0.009833 |
| }, |
| { |
| "epoch": 0.38976857490864797, |
| "grad_norm": 0.2407151460647583, |
| "learning_rate": 7.156930328406268e-05, |
| "loss": 0.6356846332550049, |
| "memory(GiB)": 71.47, |
| "step": 100, |
| "token_acc": 0.8042028604018602, |
| "train_speed(iter/s)": 0.009804 |
| }, |
| { |
| "epoch": 0.38976857490864797, |
| "eval_loss": 0.598721981048584, |
| "eval_runtime": 15.1098, |
| "eval_samples_per_second": 0.265, |
| "eval_steps_per_second": 0.132, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.4092570036540804, |
| "grad_norm": 0.2836802899837494, |
| "learning_rate": 6.861039748031351e-05, |
| "loss": 0.6598537921905517, |
| "memory(GiB)": 77.63, |
| "step": 105, |
| "token_acc": 0.7957189390414147, |
| "train_speed(iter/s)": 0.009789 |
| }, |
| { |
| "epoch": 0.4287454323995128, |
| "grad_norm": 0.26613113284111023, |
| "learning_rate": 6.557375398802123e-05, |
| "loss": 0.6290350437164307, |
| "memory(GiB)": 77.63, |
| "step": 110, |
| "token_acc": 0.8080845540515483, |
| "train_speed(iter/s)": 0.009743 |
| }, |
| { |
| "epoch": 0.4482338611449452, |
| "grad_norm": 0.3952692747116089, |
| "learning_rate": 6.247205720289907e-05, |
| "loss": 0.6128222942352295, |
| "memory(GiB)": 77.63, |
| "step": 115, |
| "token_acc": 0.8121368342872122, |
| "train_speed(iter/s)": 0.009719 |
| }, |
| { |
| "epoch": 0.46772228989037756, |
| "grad_norm": 0.29315847158432007, |
| "learning_rate": 5.9318263255459116e-05, |
| "loss": 0.6240291595458984, |
| "memory(GiB)": 77.63, |
| "step": 120, |
| "token_acc": 0.8027493895269965, |
| "train_speed(iter/s)": 0.009694 |
| }, |
| { |
| "epoch": 0.48721071863581, |
| "grad_norm": 0.2671997547149658, |
| "learning_rate": 5.6125545891822274e-05, |
| "loss": 0.6649426460266114, |
| "memory(GiB)": 77.63, |
| "step": 125, |
| "token_acc": 0.7922591927838373, |
| "train_speed(iter/s)": 0.009676 |
| }, |
| { |
| "epoch": 0.5066991473812423, |
| "grad_norm": 0.26974523067474365, |
| "learning_rate": 5.290724144552379e-05, |
| "loss": 0.6172929763793945, |
| "memory(GiB)": 77.63, |
| "step": 130, |
| "token_acc": 0.8085726351351351, |
| "train_speed(iter/s)": 0.009667 |
| }, |
| { |
| "epoch": 0.5261875761266748, |
| "grad_norm": 0.2798590660095215, |
| "learning_rate": 4.967679313017303e-05, |
| "loss": 0.6145929336547852, |
| "memory(GiB)": 77.63, |
| "step": 135, |
| "token_acc": 0.8118081180811808, |
| "train_speed(iter/s)": 0.009683 |
| }, |
| { |
| "epoch": 0.5456760048721072, |
| "grad_norm": 0.27218812704086304, |
| "learning_rate": 4.6447694885663514e-05, |
| "loss": 0.618541955947876, |
| "memory(GiB)": 77.63, |
| "step": 140, |
| "token_acc": 0.8050674649659422, |
| "train_speed(iter/s)": 0.009706 |
| }, |
| { |
| "epoch": 0.5651644336175395, |
| "grad_norm": 0.29498329758644104, |
| "learning_rate": 4.323343501249346e-05, |
| "loss": 0.6341890335083008, |
| "memory(GiB)": 77.63, |
| "step": 145, |
| "token_acc": 0.801550316061459, |
| "train_speed(iter/s)": 0.009735 |
| }, |
| { |
| "epoch": 0.584652862362972, |
| "grad_norm": 0.2799496352672577, |
| "learning_rate": 4.004743982964298e-05, |
| "loss": 0.6000133037567139, |
| "memory(GiB)": 77.63, |
| "step": 150, |
| "token_acc": 0.8125188083057479, |
| "train_speed(iter/s)": 0.009742 |
| }, |
| { |
| "epoch": 0.6041412911084044, |
| "grad_norm": 0.28971585631370544, |
| "learning_rate": 3.6903017591354706e-05, |
| "loss": 0.6476753711700439, |
| "memory(GiB)": 77.63, |
| "step": 155, |
| "token_acc": 0.8055159053833605, |
| "train_speed(iter/s)": 0.009734 |
| }, |
| { |
| "epoch": 0.6236297198538368, |
| "grad_norm": 0.2888241708278656, |
| "learning_rate": 3.381330289708396e-05, |
| "loss": 0.6604040622711181, |
| "memory(GiB)": 77.63, |
| "step": 160, |
| "token_acc": 0.7972270363951474, |
| "train_speed(iter/s)": 0.009743 |
| }, |
| { |
| "epoch": 0.6431181485992692, |
| "grad_norm": 0.31141144037246704, |
| "learning_rate": 3.079120182682412e-05, |
| "loss": 0.5943418025970459, |
| "memory(GiB)": 77.63, |
| "step": 165, |
| "token_acc": 0.8125655328607977, |
| "train_speed(iter/s)": 0.009733 |
| }, |
| { |
| "epoch": 0.6626065773447016, |
| "grad_norm": 0.28593260049819946, |
| "learning_rate": 2.7849338030983257e-05, |
| "loss": 0.6077968120574951, |
| "memory(GiB)": 77.63, |
| "step": 170, |
| "token_acc": 0.811592125382263, |
| "train_speed(iter/s)": 0.009731 |
| }, |
| { |
| "epoch": 0.682095006090134, |
| "grad_norm": 0.3223704993724823, |
| "learning_rate": 2.500000000000001e-05, |
| "loss": 0.6201191425323487, |
| "memory(GiB)": 77.63, |
| "step": 175, |
| "token_acc": 0.806087783666259, |
| "train_speed(iter/s)": 0.009728 |
| }, |
| { |
| "epoch": 0.7015834348355664, |
| "grad_norm": 0.3051086962223053, |
| "learning_rate": 2.225508973396016e-05, |
| "loss": 0.6003672122955322, |
| "memory(GiB)": 77.63, |
| "step": 180, |
| "token_acc": 0.8096822043421585, |
| "train_speed(iter/s)": 0.009738 |
| }, |
| { |
| "epoch": 0.7210718635809987, |
| "grad_norm": 0.33248814940452576, |
| "learning_rate": 1.9626073026625818e-05, |
| "loss": 0.6649184226989746, |
| "memory(GiB)": 77.63, |
| "step": 185, |
| "token_acc": 0.7927942418271744, |
| "train_speed(iter/s)": 0.009754 |
| }, |
| { |
| "epoch": 0.7405602923264312, |
| "grad_norm": 0.32766395807266235, |
| "learning_rate": 1.7123931571546827e-05, |
| "loss": 0.620220422744751, |
| "memory(GiB)": 77.63, |
| "step": 190, |
| "token_acc": 0.8047315233083407, |
| "train_speed(iter/s)": 0.009761 |
| }, |
| { |
| "epoch": 0.7600487210718636, |
| "grad_norm": 0.3414992392063141, |
| "learning_rate": 1.4759117090312197e-05, |
| "loss": 0.622031831741333, |
| "memory(GiB)": 77.63, |
| "step": 195, |
| "token_acc": 0.807596541264883, |
| "train_speed(iter/s)": 0.009765 |
| }, |
| { |
| "epoch": 0.7795371498172959, |
| "grad_norm": 0.32977092266082764, |
| "learning_rate": 1.25415076745532e-05, |
| "loss": 0.6251283168792725, |
| "memory(GiB)": 77.63, |
| "step": 200, |
| "token_acc": 0.8072234762979684, |
| "train_speed(iter/s)": 0.009771 |
| }, |
| { |
| "epoch": 0.7795371498172959, |
| "eval_loss": 0.5760576725006104, |
| "eval_runtime": 14.2053, |
| "eval_samples_per_second": 0.282, |
| "eval_steps_per_second": 0.141, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.7990255785627284, |
| "grad_norm": 0.2861402630805969, |
| "learning_rate": 1.0480366524062042e-05, |
| "loss": 0.5975393772125244, |
| "memory(GiB)": 77.63, |
| "step": 205, |
| "token_acc": 0.8152735624641129, |
| "train_speed(iter/s)": 0.009771 |
| }, |
| { |
| "epoch": 0.8185140073081608, |
| "grad_norm": 0.30515122413635254, |
| "learning_rate": 8.584303253381847e-06, |
| "loss": 0.6246855735778809, |
| "memory(GiB)": 77.63, |
| "step": 210, |
| "token_acc": 0.8046837508398119, |
| "train_speed(iter/s)": 0.009789 |
| }, |
| { |
| "epoch": 0.8380024360535931, |
| "grad_norm": 0.3188154101371765, |
| "learning_rate": 6.861237928494579e-06, |
| "loss": 0.5521413803100585, |
| "memory(GiB)": 77.63, |
| "step": 215, |
| "token_acc": 0.8276072283910298, |
| "train_speed(iter/s)": 0.009786 |
| }, |
| { |
| "epoch": 0.8574908647990256, |
| "grad_norm": 0.3368200361728668, |
| "learning_rate": 5.318367983829392e-06, |
| "loss": 0.5954047679901123, |
| "memory(GiB)": 77.63, |
| "step": 220, |
| "token_acc": 0.8130686517783292, |
| "train_speed(iter/s)": 0.009788 |
| }, |
| { |
| "epoch": 0.876979293544458, |
| "grad_norm": 0.3377918601036072, |
| "learning_rate": 3.962138157783085e-06, |
| "loss": 0.6312428951263428, |
| "memory(GiB)": 77.63, |
| "step": 225, |
| "token_acc": 0.8009544725061238, |
| "train_speed(iter/s)": 0.009783 |
| }, |
| { |
| "epoch": 0.8964677222898904, |
| "grad_norm": 0.34851112961769104, |
| "learning_rate": 2.798213572335001e-06, |
| "loss": 0.5601920127868653, |
| "memory(GiB)": 77.63, |
| "step": 230, |
| "token_acc": 0.8258236434108527, |
| "train_speed(iter/s)": 0.009778 |
| }, |
| { |
| "epoch": 0.9159561510353228, |
| "grad_norm": 0.3354482650756836, |
| "learning_rate": 1.8314560692059835e-06, |
| "loss": 0.5771200656890869, |
| "memory(GiB)": 77.63, |
| "step": 235, |
| "token_acc": 0.8201089968594125, |
| "train_speed(iter/s)": 0.009786 |
| }, |
| { |
| "epoch": 0.9354445797807551, |
| "grad_norm": 0.29914259910583496, |
| "learning_rate": 1.0659039014077944e-06, |
| "loss": 0.6100361347198486, |
| "memory(GiB)": 77.63, |
| "step": 240, |
| "token_acc": 0.8085165989634403, |
| "train_speed(iter/s)": 0.009798 |
| }, |
| { |
| "epoch": 0.9549330085261876, |
| "grad_norm": 0.36409541964530945, |
| "learning_rate": 5.047548650136513e-07, |
| "loss": 0.6194799423217774, |
| "memory(GiB)": 77.63, |
| "step": 245, |
| "token_acc": 0.807660813392776, |
| "train_speed(iter/s)": 0.009801 |
| }, |
| { |
| "epoch": 0.97442143727162, |
| "grad_norm": 0.40651243925094604, |
| "learning_rate": 1.503529416103988e-07, |
| "loss": 0.5671232223510743, |
| "memory(GiB)": 77.63, |
| "step": 250, |
| "token_acc": 0.8216745683871545, |
| "train_speed(iter/s)": 0.009809 |
| }, |
| { |
| "epoch": 0.9939098660170523, |
| "grad_norm": 0.3596270978450775, |
| "learning_rate": 4.178507228136397e-09, |
| "loss": 0.6255767822265625, |
| "memory(GiB)": 77.63, |
| "step": 255, |
| "token_acc": 0.8013068013068013, |
| "train_speed(iter/s)": 0.009788 |
| }, |
| { |
| "epoch": 0.9978075517661389, |
| "eval_loss": 0.5726350545883179, |
| "eval_runtime": 14.0103, |
| "eval_samples_per_second": 0.286, |
| "eval_steps_per_second": 0.143, |
| "step": 256 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 256, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 50, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.3517329026916086e+18, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
