| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.092896174863388, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01092896174863388, | |
| "eval_loss": 2.294912338256836, | |
| "eval_runtime": 9.8437, | |
| "eval_samples_per_second": 15.644, | |
| "eval_steps_per_second": 2.032, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03278688524590164, | |
| "grad_norm": 1.559897541999817, | |
| "learning_rate": 1.5e-05, | |
| "loss": 8.8979, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.06557377049180328, | |
| "grad_norm": 1.6194515228271484, | |
| "learning_rate": 3e-05, | |
| "loss": 9.1763, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.09836065573770492, | |
| "grad_norm": 1.691707968711853, | |
| "learning_rate": 4.5e-05, | |
| "loss": 9.2363, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.09836065573770492, | |
| "eval_loss": 2.229104518890381, | |
| "eval_runtime": 9.9458, | |
| "eval_samples_per_second": 15.484, | |
| "eval_steps_per_second": 2.011, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.13114754098360656, | |
| "grad_norm": 1.64967942237854, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 8.735, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.16393442622950818, | |
| "grad_norm": 1.896277666091919, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 8.176, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.19672131147540983, | |
| "grad_norm": 1.1657590866088867, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 7.7598, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.19672131147540983, | |
| "eval_loss": 1.973254680633545, | |
| "eval_runtime": 9.9953, | |
| "eval_samples_per_second": 15.407, | |
| "eval_steps_per_second": 2.001, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.22950819672131148, | |
| "grad_norm": 1.325817584991455, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 8.0191, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.26229508196721313, | |
| "grad_norm": 1.2216711044311523, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 7.6623, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.29508196721311475, | |
| "grad_norm": 1.25918710231781, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 7.7173, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.29508196721311475, | |
| "eval_loss": 1.913491129875183, | |
| "eval_runtime": 9.9988, | |
| "eval_samples_per_second": 15.402, | |
| "eval_steps_per_second": 2.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.32786885245901637, | |
| "grad_norm": 1.1863789558410645, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 7.6305, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.36065573770491804, | |
| "grad_norm": 1.1477130651474, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 7.6748, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.39344262295081966, | |
| "grad_norm": 1.2501188516616821, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 7.5313, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.39344262295081966, | |
| "eval_loss": 1.8740694522857666, | |
| "eval_runtime": 10.0026, | |
| "eval_samples_per_second": 15.396, | |
| "eval_steps_per_second": 1.999, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.4262295081967213, | |
| "grad_norm": 1.2684975862503052, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 7.5737, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.45901639344262296, | |
| "grad_norm": 1.303396224975586, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 7.6169, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.4918032786885246, | |
| "grad_norm": 1.1348557472229004, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 7.3236, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.4918032786885246, | |
| "eval_loss": 1.8486707210540771, | |
| "eval_runtime": 9.9999, | |
| "eval_samples_per_second": 15.4, | |
| "eval_steps_per_second": 2.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5245901639344263, | |
| "grad_norm": 1.3024663925170898, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 7.5219, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.5573770491803278, | |
| "grad_norm": 1.4016789197921753, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 7.6823, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.5901639344262295, | |
| "grad_norm": 1.232653260231018, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 7.2973, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.5901639344262295, | |
| "eval_loss": 1.835157871246338, | |
| "eval_runtime": 10.0097, | |
| "eval_samples_per_second": 15.385, | |
| "eval_steps_per_second": 1.998, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.6229508196721312, | |
| "grad_norm": 1.5029542446136475, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 7.4675, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.6557377049180327, | |
| "grad_norm": 1.2838174104690552, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 7.3278, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.6885245901639344, | |
| "grad_norm": 1.2395374774932861, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 7.4121, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.6885245901639344, | |
| "eval_loss": 1.8278913497924805, | |
| "eval_runtime": 10.0008, | |
| "eval_samples_per_second": 15.399, | |
| "eval_steps_per_second": 2.0, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.7213114754098361, | |
| "grad_norm": 1.1322687864303589, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 7.2278, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.7540983606557377, | |
| "grad_norm": 1.2498345375061035, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 7.2407, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.7868852459016393, | |
| "grad_norm": 1.4177753925323486, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 7.0907, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.7868852459016393, | |
| "eval_loss": 1.8231874704360962, | |
| "eval_runtime": 10.0062, | |
| "eval_samples_per_second": 15.39, | |
| "eval_steps_per_second": 1.999, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.819672131147541, | |
| "grad_norm": 1.3069887161254883, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 7.4671, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.8524590163934426, | |
| "grad_norm": 1.2993086576461792, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 7.5022, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.8852459016393442, | |
| "grad_norm": 1.28661048412323, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 7.0951, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.8852459016393442, | |
| "eval_loss": 1.8206969499588013, | |
| "eval_runtime": 10.005, | |
| "eval_samples_per_second": 15.392, | |
| "eval_steps_per_second": 1.999, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.9180327868852459, | |
| "grad_norm": 1.2159041166305542, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 7.4236, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.9508196721311475, | |
| "grad_norm": 1.4966245889663696, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 7.1669, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.9836065573770492, | |
| "grad_norm": 1.1949121952056885, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 7.4799, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.9836065573770492, | |
| "eval_loss": 1.8195555210113525, | |
| "eval_runtime": 10.013, | |
| "eval_samples_per_second": 15.38, | |
| "eval_steps_per_second": 1.997, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0163934426229508, | |
| "grad_norm": 1.1719740629196167, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 7.2704, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.0491803278688525, | |
| "grad_norm": 1.204927921295166, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 7.3601, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.0819672131147542, | |
| "grad_norm": 1.423250675201416, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 7.4071, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.0819672131147542, | |
| "eval_loss": 1.8196653127670288, | |
| "eval_runtime": 10.0108, | |
| "eval_samples_per_second": 15.383, | |
| "eval_steps_per_second": 1.998, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 1.3281537953680589e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |