| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.998891762098264, |
| "eval_steps": 339, |
| "global_step": 1016, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0, |
| "eval_loss": 11.850272178649902, |
| "eval_runtime": 7.1659, |
| "eval_samples_per_second": 19.956, |
| "eval_steps_per_second": 19.956, |
| "memory/device_reserved (GiB)": 3.2, |
| "memory/max_active (GiB)": 3.1, |
| "memory/max_allocated (GiB)": 3.1, |
| "step": 0 |
| }, |
| { |
| "epoch": 0.1477650535648319, |
| "grad_norm": 9.9116792678833, |
| "learning_rate": 0.00019925925947187668, |
| "loss": 7.2742, |
| "memory/device_reserved (GiB)": 7.67, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 50, |
| "tokens_per_second_per_gpu": 2082.96, |
| "total_tokens": 164572 |
| }, |
| { |
| "epoch": 0.2955301071296638, |
| "grad_norm": 12.538771629333496, |
| "learning_rate": 0.0001961624298837552, |
| "loss": 5.6623, |
| "memory/device_reserved (GiB)": 7.67, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 100, |
| "tokens_per_second_per_gpu": 1546.73, |
| "total_tokens": 287922 |
| }, |
| { |
| "epoch": 0.44329516069449576, |
| "grad_norm": 25.31467628479004, |
| "learning_rate": 0.00019072586525126637, |
| "loss": 5.3752, |
| "memory/device_reserved (GiB)": 7.67, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 150, |
| "tokens_per_second_per_gpu": 1524.55, |
| "total_tokens": 410915 |
| }, |
| { |
| "epoch": 0.5910602142593276, |
| "grad_norm": 25.589689254760742, |
| "learning_rate": 0.00018308184302213046, |
| "loss": 5.0362, |
| "memory/device_reserved (GiB)": 7.67, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 200, |
| "tokens_per_second_per_gpu": 1390.99, |
| "total_tokens": 533959 |
| }, |
| { |
| "epoch": 0.7388252678241596, |
| "grad_norm": 26.41189956665039, |
| "learning_rate": 0.00017341635045468791, |
| "loss": 4.8371, |
| "memory/device_reserved (GiB)": 7.67, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 250, |
| "tokens_per_second_per_gpu": 1552.5, |
| "total_tokens": 656838 |
| }, |
| { |
| "epoch": 0.8865903213889915, |
| "grad_norm": 23.636552810668945, |
| "learning_rate": 0.00016196455934844978, |
| "loss": 4.7248, |
| "memory/device_reserved (GiB)": 7.67, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 300, |
| "tokens_per_second_per_gpu": 1489.9, |
| "total_tokens": 779045 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 4.642312526702881, |
| "eval_runtime": 7.0402, |
| "eval_samples_per_second": 20.312, |
| "eval_steps_per_second": 20.312, |
| "memory/device_reserved (GiB)": 7.67, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 339 |
| }, |
| { |
| "epoch": 1.032508311784263, |
| "grad_norm": 29.976333618164062, |
| "learning_rate": 0.00014900510406201564, |
| "loss": 4.6412, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 350, |
| "tokens_per_second_per_gpu": 346.62, |
| "total_tokens": 945548 |
| }, |
| { |
| "epoch": 1.1802733653490949, |
| "grad_norm": 28.489376068115234, |
| "learning_rate": 0.00013485330204031937, |
| "loss": 4.4916, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 400, |
| "tokens_per_second_per_gpu": 1511.66, |
| "total_tokens": 1067762 |
| }, |
| { |
| "epoch": 1.328038418913927, |
| "grad_norm": 25.01222038269043, |
| "learning_rate": 0.0001198534818030452, |
| "loss": 4.4404, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 450, |
| "tokens_per_second_per_gpu": 1375.01, |
| "total_tokens": 1188747 |
| }, |
| { |
| "epoch": 1.4758034724787588, |
| "grad_norm": 19.93057632446289, |
| "learning_rate": 0.00010437060506248341, |
| "loss": 4.4182, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 500, |
| "tokens_per_second_per_gpu": 1540.13, |
| "total_tokens": 1312791 |
| }, |
| { |
| "epoch": 1.6235685260435906, |
| "grad_norm": 20.955829620361328, |
| "learning_rate": 8.878138681368239e-05, |
| "loss": 4.3869, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 550, |
| "tokens_per_second_per_gpu": 1544.52, |
| "total_tokens": 1437766 |
| }, |
| { |
| "epoch": 1.7713335796084226, |
| "grad_norm": 27.51922035217285, |
| "learning_rate": 7.346512945462767e-05, |
| "loss": 4.359, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 600, |
| "tokens_per_second_per_gpu": 1525.07, |
| "total_tokens": 1560469 |
| }, |
| { |
| "epoch": 1.9190986331732547, |
| "grad_norm": 18.944068908691406, |
| "learning_rate": 5.879449395213175e-05, |
| "loss": 4.3688, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 650, |
| "tokens_per_second_per_gpu": 1526.23, |
| "total_tokens": 1681459 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 4.463651657104492, |
| "eval_runtime": 6.8792, |
| "eval_samples_per_second": 20.787, |
| "eval_steps_per_second": 20.787, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 678 |
| }, |
| { |
| "epoch": 2.065016623568526, |
| "grad_norm": 18.91043472290039, |
| "learning_rate": 4.512643260086751e-05, |
| "loss": 4.294, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 700, |
| "tokens_per_second_per_gpu": 492.11, |
| "total_tokens": 1847089 |
| }, |
| { |
| "epoch": 2.212781677133358, |
| "grad_norm": 22.407855987548828, |
| "learning_rate": 3.279350399124066e-05, |
| "loss": 4.2329, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 750, |
| "tokens_per_second_per_gpu": 1508.42, |
| "total_tokens": 1969141 |
| }, |
| { |
| "epoch": 2.3605467306981898, |
| "grad_norm": 22.668535232543945, |
| "learning_rate": 2.209578150224645e-05, |
| "loss": 4.2312, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 800, |
| "tokens_per_second_per_gpu": 1531.3, |
| "total_tokens": 2090972 |
| }, |
| { |
| "epoch": 2.5083117842630216, |
| "grad_norm": 18.202590942382812, |
| "learning_rate": 1.3293552194358238e-05, |
| "loss": 4.222, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 850, |
| "tokens_per_second_per_gpu": 1544.96, |
| "total_tokens": 2213097 |
| }, |
| { |
| "epoch": 2.656076837827854, |
| "grad_norm": 18.34627914428711, |
| "learning_rate": 6.600983746212319e-06, |
| "loss": 4.2271, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 900, |
| "tokens_per_second_per_gpu": 1515.29, |
| "total_tokens": 2335026 |
| }, |
| { |
| "epoch": 2.8038418913926857, |
| "grad_norm": 21.71125030517578, |
| "learning_rate": 2.1809135253115565e-06, |
| "loss": 4.2248, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 950, |
| "tokens_per_second_per_gpu": 1450.1, |
| "total_tokens": 2458575 |
| }, |
| { |
| "epoch": 2.9516069449575175, |
| "grad_norm": 16.12077522277832, |
| "learning_rate": 1.4088658024622448e-07, |
| "loss": 4.2186, |
| "memory/device_reserved (GiB)": 7.62, |
| "memory/max_active (GiB)": 7.2, |
| "memory/max_allocated (GiB)": 7.2, |
| "step": 1000, |
| "tokens_per_second_per_gpu": 1481.33, |
| "total_tokens": 2582586 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 1016, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 200, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.7850881269039104e+16, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
