| { |
| "best_metric": 0.65715516, |
| "best_model_checkpoint": "Qwen2.5-7B-Instruct_lora/checkpoint-128", |
| "epoch": 0.997564539698003, |
| "eval_steps": 100, |
| "global_step": 128, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.007793472966390648, |
| "grad_norm": 0.4758923351764679, |
| "learning_rate": 1.4285714285714285e-05, |
| "loss": 0.82492995262146, |
| "memory(GiB)": 77.98, |
| "step": 1, |
| "token_acc": 0.7762530413625304, |
| "train_speed(iter/s)": 0.01272 |
| }, |
| { |
| "epoch": 0.03896736483195324, |
| "grad_norm": 0.5201897621154785, |
| "learning_rate": 7.142857142857143e-05, |
| "loss": 0.9614015817642212, |
| "memory(GiB)": 77.98, |
| "step": 5, |
| "token_acc": 0.7459884860362568, |
| "train_speed(iter/s)": 0.013169 |
| }, |
| { |
| "epoch": 0.07793472966390648, |
| "grad_norm": 0.3151409327983856, |
| "learning_rate": 9.984840253435568e-05, |
| "loss": 0.8129005432128906, |
| "memory(GiB)": 77.98, |
| "step": 10, |
| "token_acc": 0.7703662699511025, |
| "train_speed(iter/s)": 0.013385 |
| }, |
| { |
| "epoch": 0.11690209449585971, |
| "grad_norm": 0.22351491451263428, |
| "learning_rate": 9.892530053338909e-05, |
| "loss": 0.8129632949829102, |
| "memory(GiB)": 78.1, |
| "step": 15, |
| "token_acc": 0.7659050029978016, |
| "train_speed(iter/s)": 0.013457 |
| }, |
| { |
| "epoch": 0.15586945932781296, |
| "grad_norm": 0.20141544938087463, |
| "learning_rate": 9.71788333255991e-05, |
| "loss": 0.737860107421875, |
| "memory(GiB)": 78.1, |
| "step": 20, |
| "token_acc": 0.780184425765231, |
| "train_speed(iter/s)": 0.013638 |
| }, |
| { |
| "epoch": 0.1948368241597662, |
| "grad_norm": 0.19052022695541382, |
| "learning_rate": 9.46383922548932e-05, |
| "loss": 0.7453306198120118, |
| "memory(GiB)": 78.1, |
| "step": 25, |
| "token_acc": 0.7835497835497836, |
| "train_speed(iter/s)": 0.013751 |
| }, |
| { |
| "epoch": 0.23380418899171942, |
| "grad_norm": 0.19930361211299896, |
| "learning_rate": 9.134673047280645e-05, |
| "loss": 0.7681457042694092, |
| "memory(GiB)": 78.1, |
| "step": 30, |
| "token_acc": 0.7709214619843775, |
| "train_speed(iter/s)": 0.013864 |
| }, |
| { |
| "epoch": 0.2727715538236727, |
| "grad_norm": 0.1565362960100174, |
| "learning_rate": 8.735924344455732e-05, |
| "loss": 0.684519624710083, |
| "memory(GiB)": 78.1, |
| "step": 35, |
| "token_acc": 0.7952077263187921, |
| "train_speed(iter/s)": 0.013849 |
| }, |
| { |
| "epoch": 0.3117389186556259, |
| "grad_norm": 0.1997351050376892, |
| "learning_rate": 8.274303669726426e-05, |
| "loss": 0.7397434711456299, |
| "memory(GiB)": 78.1, |
| "step": 40, |
| "token_acc": 0.7784489429149091, |
| "train_speed(iter/s)": 0.013865 |
| }, |
| { |
| "epoch": 0.35070628348757915, |
| "grad_norm": 0.23139981925487518, |
| "learning_rate": 7.757579649921354e-05, |
| "loss": 0.7341322422027587, |
| "memory(GiB)": 78.1, |
| "step": 45, |
| "token_acc": 0.7790655206836941, |
| "train_speed(iter/s)": 0.013866 |
| }, |
| { |
| "epoch": 0.3896736483195324, |
| "grad_norm": 0.19183649122714996, |
| "learning_rate": 7.19444824755478e-05, |
| "loss": 0.7072115421295166, |
| "memory(GiB)": 78.1, |
| "step": 50, |
| "token_acc": 0.7878044975372752, |
| "train_speed(iter/s)": 0.013846 |
| }, |
| { |
| "epoch": 0.4286410131514856, |
| "grad_norm": 0.22789426147937775, |
| "learning_rate": 6.594386416238094e-05, |
| "loss": 0.7093594074249268, |
| "memory(GiB)": 78.1, |
| "step": 55, |
| "token_acc": 0.7885790362075934, |
| "train_speed(iter/s)": 0.013798 |
| }, |
| { |
| "epoch": 0.46760837798343885, |
| "grad_norm": 0.21530242264270782, |
| "learning_rate": 5.967492612770999e-05, |
| "loss": 0.6775043964385986, |
| "memory(GiB)": 78.1, |
| "step": 60, |
| "token_acc": 0.7929026953758639, |
| "train_speed(iter/s)": 0.01371 |
| }, |
| { |
| "epoch": 0.5065757428153921, |
| "grad_norm": 0.20659631490707397, |
| "learning_rate": 5.324316849938715e-05, |
| "loss": 0.7007603645324707, |
| "memory(GiB)": 78.1, |
| "step": 65, |
| "token_acc": 0.787200775176727, |
| "train_speed(iter/s)": 0.013663 |
| }, |
| { |
| "epoch": 0.5455431076473454, |
| "grad_norm": 0.2197224497795105, |
| "learning_rate": 4.675683150061285e-05, |
| "loss": 0.6781778335571289, |
| "memory(GiB)": 78.1, |
| "step": 70, |
| "token_acc": 0.7972211791212918, |
| "train_speed(iter/s)": 0.013694 |
| }, |
| { |
| "epoch": 0.5845104724792985, |
| "grad_norm": 0.24823956191539764, |
| "learning_rate": 4.032507387229002e-05, |
| "loss": 0.6829503536224365, |
| "memory(GiB)": 78.1, |
| "step": 75, |
| "token_acc": 0.7936122659873591, |
| "train_speed(iter/s)": 0.013751 |
| }, |
| { |
| "epoch": 0.6234778373112518, |
| "grad_norm": 0.20789459347724915, |
| "learning_rate": 3.4056135837619074e-05, |
| "loss": 0.7214729309082031, |
| "memory(GiB)": 78.1, |
| "step": 80, |
| "token_acc": 0.7854131534569984, |
| "train_speed(iter/s)": 0.013775 |
| }, |
| { |
| "epoch": 0.662445202143205, |
| "grad_norm": 0.22102876007556915, |
| "learning_rate": 2.805551752445222e-05, |
| "loss": 0.6610813617706299, |
| "memory(GiB)": 78.1, |
| "step": 85, |
| "token_acc": 0.7997587724196467, |
| "train_speed(iter/s)": 0.013766 |
| }, |
| { |
| "epoch": 0.7014125669751583, |
| "grad_norm": 0.24639996886253357, |
| "learning_rate": 2.2424203500786474e-05, |
| "loss": 0.6775108814239502, |
| "memory(GiB)": 78.1, |
| "step": 90, |
| "token_acc": 0.7919411605450771, |
| "train_speed(iter/s)": 0.013768 |
| }, |
| { |
| "epoch": 0.7403799318071116, |
| "grad_norm": 0.26909467577934265, |
| "learning_rate": 1.725696330273575e-05, |
| "loss": 0.7113357543945312, |
| "memory(GiB)": 78.1, |
| "step": 95, |
| "token_acc": 0.7827240947736395, |
| "train_speed(iter/s)": 0.013806 |
| }, |
| { |
| "epoch": 0.7793472966390648, |
| "grad_norm": 0.2722283601760864, |
| "learning_rate": 1.2640756555442684e-05, |
| "loss": 0.6915526390075684, |
| "memory(GiB)": 78.1, |
| "step": 100, |
| "token_acc": 0.7925359976491331, |
| "train_speed(iter/s)": 0.013811 |
| }, |
| { |
| "epoch": 0.7793472966390648, |
| "eval_loss": 0.6591519713401794, |
| "eval_runtime": 3.54, |
| "eval_samples_per_second": 1.13, |
| "eval_steps_per_second": 0.565, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.8183146614710181, |
| "grad_norm": 0.24852623045444489, |
| "learning_rate": 8.65326952719357e-06, |
| "loss": 0.6732324600219727, |
| "memory(GiB)": 78.1, |
| "step": 105, |
| "token_acc": 0.7969924812030075, |
| "train_speed(iter/s)": 0.013826 |
| }, |
| { |
| "epoch": 0.8572820263029712, |
| "grad_norm": 0.2657414972782135, |
| "learning_rate": 5.361607745106817e-06, |
| "loss": 0.6405797481536866, |
| "memory(GiB)": 78.1, |
| "step": 110, |
| "token_acc": 0.8046042213240682, |
| "train_speed(iter/s)": 0.013822 |
| }, |
| { |
| "epoch": 0.8962493911349245, |
| "grad_norm": 0.23174554109573364, |
| "learning_rate": 2.8211666744009047e-06, |
| "loss": 0.6503153800964355, |
| "memory(GiB)": 78.1, |
| "step": 115, |
| "token_acc": 0.7991786632970802, |
| "train_speed(iter/s)": 0.013813 |
| }, |
| { |
| "epoch": 0.9352167559668777, |
| "grad_norm": 0.24391594529151917, |
| "learning_rate": 1.0746994666109234e-06, |
| "loss": 0.6571870803833008, |
| "memory(GiB)": 78.1, |
| "step": 120, |
| "token_acc": 0.799995356288746, |
| "train_speed(iter/s)": 0.013833 |
| }, |
| { |
| "epoch": 0.974184120798831, |
| "grad_norm": 0.25083476305007935, |
| "learning_rate": 1.5159746564433198e-07, |
| "loss": 0.6554246425628663, |
| "memory(GiB)": 78.1, |
| "step": 125, |
| "token_acc": 0.7995940460081191, |
| "train_speed(iter/s)": 0.013845 |
| }, |
| { |
| "epoch": 0.997564539698003, |
| "eval_loss": 0.6571551561355591, |
| "eval_runtime": 3.9227, |
| "eval_samples_per_second": 1.02, |
| "eval_steps_per_second": 0.51, |
| "step": 128 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 128, |
| "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": 8.054032501501809e+17, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
