| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.975609756097561, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00975609756097561, | |
| "eval_loss": 3.28096079826355, | |
| "eval_runtime": 4.8657, | |
| "eval_samples_per_second": 35.555, | |
| "eval_steps_per_second": 4.521, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02926829268292683, | |
| "grad_norm": 0.18293029069900513, | |
| "learning_rate": 1.5e-05, | |
| "loss": 3.1125, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.05853658536585366, | |
| "grad_norm": 0.3248441815376282, | |
| "learning_rate": 3e-05, | |
| "loss": 3.427, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.08780487804878048, | |
| "grad_norm": 0.2983645498752594, | |
| "learning_rate": 4.5e-05, | |
| "loss": 3.3708, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.08780487804878048, | |
| "eval_loss": 3.2691893577575684, | |
| "eval_runtime": 4.8711, | |
| "eval_samples_per_second": 35.515, | |
| "eval_steps_per_second": 4.516, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.11707317073170732, | |
| "grad_norm": 0.17559802532196045, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 3.0003, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.14634146341463414, | |
| "grad_norm": 0.4407211244106293, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 3.3846, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.17560975609756097, | |
| "grad_norm": 0.24928590655326843, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 2.9842, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.17560975609756097, | |
| "eval_loss": 3.209193229675293, | |
| "eval_runtime": 4.8642, | |
| "eval_samples_per_second": 35.566, | |
| "eval_steps_per_second": 4.523, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2048780487804878, | |
| "grad_norm": 0.3672405481338501, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 3.1769, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.23414634146341465, | |
| "grad_norm": 0.3140566647052765, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 2.8991, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2634146341463415, | |
| "grad_norm": 0.23150236904621124, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 2.7151, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2634146341463415, | |
| "eval_loss": 3.1205644607543945, | |
| "eval_runtime": 4.8858, | |
| "eval_samples_per_second": 35.409, | |
| "eval_steps_per_second": 4.503, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2926829268292683, | |
| "grad_norm": 0.1518729329109192, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 2.7446, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.32195121951219513, | |
| "grad_norm": 0.16655544936656952, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 2.7515, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.35121951219512193, | |
| "grad_norm": 0.2755165696144104, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 3.1518, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.35121951219512193, | |
| "eval_loss": 3.039105176925659, | |
| "eval_runtime": 4.9041, | |
| "eval_samples_per_second": 35.277, | |
| "eval_steps_per_second": 4.486, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.3804878048780488, | |
| "grad_norm": 0.18405312299728394, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 2.7984, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.4097560975609756, | |
| "grad_norm": 0.38106638193130493, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 3.1116, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.43902439024390244, | |
| "grad_norm": 0.21435827016830444, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 2.7819, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.43902439024390244, | |
| "eval_loss": 2.9602138996124268, | |
| "eval_runtime": 4.8684, | |
| "eval_samples_per_second": 35.535, | |
| "eval_steps_per_second": 4.519, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.4682926829268293, | |
| "grad_norm": 0.234124556183815, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 3.0273, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.4975609756097561, | |
| "grad_norm": 0.24075672030448914, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 2.961, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.526829268292683, | |
| "grad_norm": 0.3469984233379364, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 2.7697, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.526829268292683, | |
| "eval_loss": 2.9034578800201416, | |
| "eval_runtime": 4.9077, | |
| "eval_samples_per_second": 35.251, | |
| "eval_steps_per_second": 4.483, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.5560975609756098, | |
| "grad_norm": 0.28564637899398804, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 2.7876, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.5853658536585366, | |
| "grad_norm": 0.2155310958623886, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 2.8619, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.6146341463414634, | |
| "grad_norm": 0.2307025045156479, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 2.5967, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.6146341463414634, | |
| "eval_loss": 2.866178512573242, | |
| "eval_runtime": 4.8788, | |
| "eval_samples_per_second": 35.46, | |
| "eval_steps_per_second": 4.509, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.6439024390243903, | |
| "grad_norm": 0.24308180809020996, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 2.6941, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.6731707317073171, | |
| "grad_norm": 0.22699187695980072, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 3.0174, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.7024390243902439, | |
| "grad_norm": 0.20896831154823303, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 2.8152, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.7024390243902439, | |
| "eval_loss": 2.84454083442688, | |
| "eval_runtime": 4.8793, | |
| "eval_samples_per_second": 35.456, | |
| "eval_steps_per_second": 4.509, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.7317073170731707, | |
| "grad_norm": 0.2799305021762848, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 2.6651, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.7609756097560976, | |
| "grad_norm": 0.24884013831615448, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 3.0048, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.7902439024390244, | |
| "grad_norm": 0.3715744614601135, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 2.8045, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.7902439024390244, | |
| "eval_loss": 2.8328237533569336, | |
| "eval_runtime": 4.8668, | |
| "eval_samples_per_second": 35.547, | |
| "eval_steps_per_second": 4.52, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.8195121951219512, | |
| "grad_norm": 0.3776783049106598, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 2.6349, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.848780487804878, | |
| "grad_norm": 0.21473738551139832, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 2.6227, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.8780487804878049, | |
| "grad_norm": 0.21475864946842194, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 2.8433, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8780487804878049, | |
| "eval_loss": 2.828202247619629, | |
| "eval_runtime": 4.8709, | |
| "eval_samples_per_second": 35.517, | |
| "eval_steps_per_second": 4.517, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.9073170731707317, | |
| "grad_norm": 0.30402669310569763, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 2.8122, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.9365853658536586, | |
| "grad_norm": 0.6135450005531311, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 2.7137, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.9658536585365853, | |
| "grad_norm": 0.3046834468841553, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 2.7436, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.9658536585365853, | |
| "eval_loss": 2.82761549949646, | |
| "eval_runtime": 4.8837, | |
| "eval_samples_per_second": 35.424, | |
| "eval_steps_per_second": 4.505, | |
| "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": 3.40250018709504e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |