| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03751641343087601, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00037516413430876007, | |
| "eval_loss": NaN, | |
| "eval_runtime": 414.1988, | |
| "eval_samples_per_second": 10.84, | |
| "eval_steps_per_second": 1.357, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0011254924029262803, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0022509848058525606, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0033764772087788407, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0033764772087788407, | |
| "eval_loss": NaN, | |
| "eval_runtime": 414.7893, | |
| "eval_samples_per_second": 10.825, | |
| "eval_steps_per_second": 1.355, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004501969611705121, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.005627462014631401, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.006752954417557681, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.006752954417557681, | |
| "eval_loss": NaN, | |
| "eval_runtime": 414.8491, | |
| "eval_samples_per_second": 10.823, | |
| "eval_steps_per_second": 1.355, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.007878446820483961, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 0.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.009003939223410242, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 0.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.010129431626336522, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.010129431626336522, | |
| "eval_loss": NaN, | |
| "eval_runtime": 414.7432, | |
| "eval_samples_per_second": 10.826, | |
| "eval_steps_per_second": 1.355, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.011254924029262802, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.012380416432189083, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 0.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.013505908835115363, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.013505908835115363, | |
| "eval_loss": NaN, | |
| "eval_runtime": 414.8341, | |
| "eval_samples_per_second": 10.824, | |
| "eval_steps_per_second": 1.355, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.014631401238041642, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.015756893640967922, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.016882386043894203, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.016882386043894203, | |
| "eval_loss": NaN, | |
| "eval_runtime": 414.8566, | |
| "eval_samples_per_second": 10.823, | |
| "eval_steps_per_second": 1.355, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.018007878446820485, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.019133370849746763, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.0, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.020258863252673044, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.0, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.020258863252673044, | |
| "eval_loss": NaN, | |
| "eval_runtime": 414.8775, | |
| "eval_samples_per_second": 10.822, | |
| "eval_steps_per_second": 1.355, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.021384355655599326, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.0, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.022509848058525603, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.023635340461451885, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.0, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.023635340461451885, | |
| "eval_loss": NaN, | |
| "eval_runtime": 414.7903, | |
| "eval_samples_per_second": 10.825, | |
| "eval_steps_per_second": 1.355, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.024760832864378166, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.0, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.025886325267304444, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.0, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.027011817670230726, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.0, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.027011817670230726, | |
| "eval_loss": NaN, | |
| "eval_runtime": 414.8179, | |
| "eval_samples_per_second": 10.824, | |
| "eval_steps_per_second": 1.355, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.028137310073157007, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.029262802476083285, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.0, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.030388294879009566, | |
| "grad_norm": NaN, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.0, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.030388294879009566, | |
| "eval_loss": NaN, | |
| "eval_runtime": 414.7899, | |
| "eval_samples_per_second": 10.825, | |
| "eval_steps_per_second": 1.355, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.031513787281935844, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.0, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.032639279684862126, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.0, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.03376477208778841, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.03376477208778841, | |
| "eval_loss": NaN, | |
| "eval_runtime": 415.1158, | |
| "eval_samples_per_second": 10.816, | |
| "eval_steps_per_second": 1.354, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.03489026449071469, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.0, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.03601575689364097, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.0, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.03714124929656725, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.0, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.03714124929656725, | |
| "eval_loss": NaN, | |
| "eval_runtime": 414.9211, | |
| "eval_samples_per_second": 10.821, | |
| "eval_steps_per_second": 1.354, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.909031571428475e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |