| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.23617619493908154, | |
| "eval_steps": 9, | |
| "global_step": 63, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0037488284910965324, | |
| "eval_loss": 2.055473566055298, | |
| "eval_runtime": 12.7703, | |
| "eval_samples_per_second": 35.238, | |
| "eval_steps_per_second": 4.463, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011246485473289597, | |
| "grad_norm": 0.1467806100845337, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.9063, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.022492970946579195, | |
| "grad_norm": 0.18374566733837128, | |
| "learning_rate": 3e-05, | |
| "loss": 1.9823, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.033739456419868794, | |
| "grad_norm": 0.2081272304058075, | |
| "learning_rate": 4.5e-05, | |
| "loss": 2.1133, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.033739456419868794, | |
| "eval_loss": 2.050758123397827, | |
| "eval_runtime": 12.8585, | |
| "eval_samples_per_second": 34.996, | |
| "eval_steps_per_second": 4.433, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.04498594189315839, | |
| "grad_norm": 0.2559334337711334, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 2.0606, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.056232427366447985, | |
| "grad_norm": 0.15697269141674042, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 2.0357, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.06747891283973759, | |
| "grad_norm": 0.16153453290462494, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 2.005, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.06747891283973759, | |
| "eval_loss": 2.0186078548431396, | |
| "eval_runtime": 12.902, | |
| "eval_samples_per_second": 34.878, | |
| "eval_steps_per_second": 4.418, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07872539831302718, | |
| "grad_norm": 0.15399610996246338, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 1.9246, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.08997188378631678, | |
| "grad_norm": 0.18686576187610626, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 1.9773, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.10121836925960637, | |
| "grad_norm": 0.1838475912809372, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 2.0225, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10121836925960637, | |
| "eval_loss": 1.988869071006775, | |
| "eval_runtime": 12.9477, | |
| "eval_samples_per_second": 34.755, | |
| "eval_steps_per_second": 4.402, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.11246485473289597, | |
| "grad_norm": 0.19368326663970947, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 1.9612, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12371134020618557, | |
| "grad_norm": 0.17149889469146729, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 2.022, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.13495782567947517, | |
| "grad_norm": 0.18569713830947876, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 1.9794, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.13495782567947517, | |
| "eval_loss": 1.9658715724945068, | |
| "eval_runtime": 12.9648, | |
| "eval_samples_per_second": 34.709, | |
| "eval_steps_per_second": 4.397, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.14620431115276475, | |
| "grad_norm": 0.1827002316713333, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 1.889, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.15745079662605435, | |
| "grad_norm": 0.23876498639583588, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 2.0144, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.16869728209934395, | |
| "grad_norm": 0.22354301810264587, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 1.9035, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.16869728209934395, | |
| "eval_loss": 1.9489043951034546, | |
| "eval_runtime": 12.9724, | |
| "eval_samples_per_second": 34.689, | |
| "eval_steps_per_second": 4.394, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.17994376757263356, | |
| "grad_norm": 0.22030919790267944, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 2.0074, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.19119025304592316, | |
| "grad_norm": 0.20402924716472626, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 1.9598, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.20243673851921273, | |
| "grad_norm": 0.21052472293376923, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 1.9229, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.20243673851921273, | |
| "eval_loss": 1.9371302127838135, | |
| "eval_runtime": 12.9738, | |
| "eval_samples_per_second": 34.685, | |
| "eval_steps_per_second": 4.393, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.21368322399250234, | |
| "grad_norm": 0.20837906002998352, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 1.818, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.22492970946579194, | |
| "grad_norm": 0.22922676801681519, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 2.0037, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.23617619493908154, | |
| "grad_norm": 0.1948896199464798, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 1.8037, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.23617619493908154, | |
| "eval_loss": 1.9288066625595093, | |
| "eval_runtime": 13.0065, | |
| "eval_samples_per_second": 34.598, | |
| "eval_steps_per_second": 4.382, | |
| "step": 63 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.0575118778433536e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |