| { | |
| "best_metric": 0.5390731692314148, | |
| "best_model_checkpoint": "/ML-A100/team/mm/eamon/self_instruction/seed_ppl/qwen14B_models/qwen_14B_d1_iter2_model/checkpoint-200", | |
| "epoch": 2.0, | |
| "eval_steps": 100, | |
| "global_step": 338, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05917159763313609, | |
| "grad_norm": 1.5490762637870956, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.39, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11834319526627218, | |
| "grad_norm": 2.180427642573537, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2536, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.17751479289940827, | |
| "grad_norm": 1.4938352528249466, | |
| "learning_rate": 4.98781004037916e-05, | |
| "loss": 0.6182, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.23668639053254437, | |
| "grad_norm": 0.8266774666049563, | |
| "learning_rate": 4.951359037609088e-05, | |
| "loss": 0.5871, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2958579881656805, | |
| "grad_norm": 0.44331898296236955, | |
| "learning_rate": 4.891002460691306e-05, | |
| "loss": 0.6622, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.35502958579881655, | |
| "grad_norm": 0.7348699943282603, | |
| "learning_rate": 4.807328905014201e-05, | |
| "loss": 0.5239, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.41420118343195267, | |
| "grad_norm": 1.0233758530765529, | |
| "learning_rate": 4.7011543523897996e-05, | |
| "loss": 0.6031, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.47337278106508873, | |
| "grad_norm": 0.44566662901156306, | |
| "learning_rate": 4.573514213625505e-05, | |
| "loss": 0.5342, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5325443786982249, | |
| "grad_norm": 0.32376173126197005, | |
| "learning_rate": 4.425653231231344e-05, | |
| "loss": 0.5295, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.591715976331361, | |
| "grad_norm": 2.5050276715521953, | |
| "learning_rate": 4.259013340731224e-05, | |
| "loss": 0.5923, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.591715976331361, | |
| "eval_loss": 0.5548861622810364, | |
| "eval_runtime": 20.5369, | |
| "eval_samples_per_second": 14.608, | |
| "eval_steps_per_second": 1.85, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.650887573964497, | |
| "grad_norm": 0.6939641269685894, | |
| "learning_rate": 4.075219608954278e-05, | |
| "loss": 0.5283, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.7100591715976331, | |
| "grad_norm": 0.3223403825561066, | |
| "learning_rate": 3.876064386435646e-05, | |
| "loss": 0.5123, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 1.242304136081819, | |
| "learning_rate": 3.663489828471953e-05, | |
| "loss": 0.5148, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.8284023668639053, | |
| "grad_norm": 0.6162370328893566, | |
| "learning_rate": 3.4395689552855955e-05, | |
| "loss": 0.5706, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.8875739644970414, | |
| "grad_norm": 2.854595963693751, | |
| "learning_rate": 3.206485435998498e-05, | |
| "loss": 0.5443, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.9467455621301775, | |
| "grad_norm": 0.836450705626818, | |
| "learning_rate": 2.9665122935613727e-05, | |
| "loss": 0.5345, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0059171597633136, | |
| "grad_norm": 0.6854327873281724, | |
| "learning_rate": 2.7219897383073373e-05, | |
| "loss": 0.4759, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.0650887573964498, | |
| "grad_norm": 0.3289222105005388, | |
| "learning_rate": 2.475302346296336e-05, | |
| "loss": 0.4207, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.1242603550295858, | |
| "grad_norm": 0.6402929677890956, | |
| "learning_rate": 2.2288558050064367e-05, | |
| "loss": 0.4453, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.183431952662722, | |
| "grad_norm": 0.4560254960950853, | |
| "learning_rate": 1.9850534531472546e-05, | |
| "loss": 0.485, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.183431952662722, | |
| "eval_loss": 0.5390731692314148, | |
| "eval_runtime": 19.171, | |
| "eval_samples_per_second": 15.649, | |
| "eval_steps_per_second": 1.982, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.242603550295858, | |
| "grad_norm": 0.4429967217295556, | |
| "learning_rate": 1.746272843378493e-05, | |
| "loss": 0.448, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.301775147928994, | |
| "grad_norm": 0.5801927088218751, | |
| "learning_rate": 1.5148425564932084e-05, | |
| "loss": 0.4257, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.3609467455621302, | |
| "grad_norm": 1.6461621420887593, | |
| "learning_rate": 1.2930194931731382e-05, | |
| "loss": 0.5102, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.4201183431952662, | |
| "grad_norm": 1.4629083533873057, | |
| "learning_rate": 1.0829668647661559e-05, | |
| "loss": 0.4632, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.4792899408284024, | |
| "grad_norm": 1.6289265138854685, | |
| "learning_rate": 8.867330977190877e-06, | |
| "loss": 0.5108, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.6682669143854097, | |
| "learning_rate": 7.062318573891716e-06, | |
| "loss": 0.4304, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.5976331360946747, | |
| "grad_norm": 0.5194393122476173, | |
| "learning_rate": 5.4322338604131715e-06, | |
| "loss": 0.4281, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.6568047337278107, | |
| "grad_norm": 1.0124557614072613, | |
| "learning_rate": 3.992973370223896e-06, | |
| "loss": 0.4529, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.7159763313609466, | |
| "grad_norm": 0.42873082380830013, | |
| "learning_rate": 2.75857272513132e-06, | |
| "loss": 0.4698, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.7751479289940828, | |
| "grad_norm": 0.9876326963962394, | |
| "learning_rate": 1.7410697603511383e-06, | |
| "loss": 0.4405, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.7751479289940828, | |
| "eval_loss": 0.5398803353309631, | |
| "eval_runtime": 18.7716, | |
| "eval_samples_per_second": 15.982, | |
| "eval_steps_per_second": 2.024, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.834319526627219, | |
| "grad_norm": 0.9194473587313593, | |
| "learning_rate": 9.503871319271551e-07, | |
| "loss": 0.4825, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.893491124260355, | |
| "grad_norm": 1.330707485363624, | |
| "learning_rate": 3.9423555131007925e-07, | |
| "loss": 0.5166, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.952662721893491, | |
| "grad_norm": 0.3585144579596122, | |
| "learning_rate": 7.803859074854425e-08, | |
| "loss": 0.4212, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 338, | |
| "total_flos": 93650261901312.0, | |
| "train_loss": 0.5521188734551153, | |
| "train_runtime": 1213.4234, | |
| "train_samples_per_second": 4.45, | |
| "train_steps_per_second": 0.279 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 338, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 93650261901312.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |