| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5695076132094127, | |
| "eval_steps": 8, | |
| "global_step": 90, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.006327862368993475, | |
| "eval_loss": 4.022780895233154, | |
| "eval_runtime": 85.0755, | |
| "eval_samples_per_second": 6.265, | |
| "eval_steps_per_second": 3.138, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.018983587106980424, | |
| "grad_norm": 1.016599416732788, | |
| "learning_rate": 3e-05, | |
| "loss": 3.1283, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.03796717421396085, | |
| "grad_norm": 1.0906339883804321, | |
| "learning_rate": 6e-05, | |
| "loss": 2.94, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0506228989519478, | |
| "eval_loss": 3.2956628799438477, | |
| "eval_runtime": 85.1034, | |
| "eval_samples_per_second": 6.263, | |
| "eval_steps_per_second": 3.137, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.05695076132094127, | |
| "grad_norm": 1.010114312171936, | |
| "learning_rate": 9e-05, | |
| "loss": 2.6706, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0759343484279217, | |
| "grad_norm": 1.208998441696167, | |
| "learning_rate": 0.00012, | |
| "loss": 2.3263, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.09491793553490212, | |
| "grad_norm": 1.0851699113845825, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.8705, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1012457979038956, | |
| "eval_loss": 1.7172771692276, | |
| "eval_runtime": 85.0679, | |
| "eval_samples_per_second": 6.266, | |
| "eval_steps_per_second": 3.139, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.11390152264188254, | |
| "grad_norm": 1.24854576587677, | |
| "learning_rate": 0.00018, | |
| "loss": 1.6898, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.13288510974886297, | |
| "grad_norm": 0.9713943600654602, | |
| "learning_rate": 0.00019989930665413147, | |
| "loss": 1.4369, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.1518686968558434, | |
| "grad_norm": 0.825309157371521, | |
| "learning_rate": 0.00019839295885986296, | |
| "loss": 1.1792, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.1518686968558434, | |
| "eval_loss": 1.2726812362670898, | |
| "eval_runtime": 85.0943, | |
| "eval_samples_per_second": 6.264, | |
| "eval_steps_per_second": 3.138, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.17085228396282381, | |
| "grad_norm": 1.0981557369232178, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 1.2949, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.18983587106980424, | |
| "grad_norm": 0.9579986333847046, | |
| "learning_rate": 0.0001900968867902419, | |
| "loss": 1.4622, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2024915958077912, | |
| "eval_loss": 1.1003365516662598, | |
| "eval_runtime": 85.0901, | |
| "eval_samples_per_second": 6.264, | |
| "eval_steps_per_second": 3.138, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.20881945817678466, | |
| "grad_norm": 0.5367055535316467, | |
| "learning_rate": 0.00018345732537213027, | |
| "loss": 0.948, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.22780304528376508, | |
| "grad_norm": 0.6562434434890747, | |
| "learning_rate": 0.00017530714660036112, | |
| "loss": 1.0054, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2467866323907455, | |
| "grad_norm": 0.795328676700592, | |
| "learning_rate": 0.00016579387259397127, | |
| "loss": 1.0216, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.25311449475973896, | |
| "eval_loss": 0.9722965955734253, | |
| "eval_runtime": 85.1078, | |
| "eval_samples_per_second": 6.263, | |
| "eval_steps_per_second": 3.137, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.26577021949772595, | |
| "grad_norm": 0.5210263133049011, | |
| "learning_rate": 0.00015508969814521025, | |
| "loss": 0.9488, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.28475380660470634, | |
| "grad_norm": 0.614801824092865, | |
| "learning_rate": 0.00014338837391175582, | |
| "loss": 0.6938, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.3037373937116868, | |
| "grad_norm": 0.8110975027084351, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 1.1436, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.3037373937116868, | |
| "eval_loss": 0.8694770336151123, | |
| "eval_runtime": 85.0621, | |
| "eval_samples_per_second": 6.266, | |
| "eval_steps_per_second": 3.139, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.3227209808186672, | |
| "grad_norm": 0.5281067490577698, | |
| "learning_rate": 0.00011785568947986367, | |
| "loss": 0.7773, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.34170456792564763, | |
| "grad_norm": 0.6224583387374878, | |
| "learning_rate": 0.00010448648303505151, | |
| "loss": 0.8854, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.35436029266363456, | |
| "eval_loss": 0.7901315689086914, | |
| "eval_runtime": 85.0645, | |
| "eval_samples_per_second": 6.266, | |
| "eval_steps_per_second": 3.139, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.360688155032628, | |
| "grad_norm": 0.45073410868644714, | |
| "learning_rate": 9.103606910965666e-05, | |
| "loss": 0.677, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.37967174213960847, | |
| "grad_norm": 1.052748680114746, | |
| "learning_rate": 7.774790660436858e-05, | |
| "loss": 0.8509, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.39865532924658886, | |
| "grad_norm": 0.6174542307853699, | |
| "learning_rate": 6.486251759186572e-05, | |
| "loss": 0.6693, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.4049831916155824, | |
| "eval_loss": 0.7380265593528748, | |
| "eval_runtime": 85.1209, | |
| "eval_samples_per_second": 6.262, | |
| "eval_steps_per_second": 3.137, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.4176389163535693, | |
| "grad_norm": 0.6812132000923157, | |
| "learning_rate": 5.261313375270014e-05, | |
| "loss": 0.6868, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.43662250346054976, | |
| "grad_norm": 0.6077867150306702, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.7692, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.45560609056753015, | |
| "grad_norm": 0.6931792497634888, | |
| "learning_rate": 3.089373510131354e-05, | |
| "loss": 0.7443, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.45560609056753015, | |
| "eval_loss": 0.6958061456680298, | |
| "eval_runtime": 85.0295, | |
| "eval_samples_per_second": 6.268, | |
| "eval_steps_per_second": 3.14, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.4745896776745106, | |
| "grad_norm": 1.234330415725708, | |
| "learning_rate": 2.181685175319702e-05, | |
| "loss": 0.6574, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.493573264781491, | |
| "grad_norm": 0.7910563349723816, | |
| "learning_rate": 1.415512063981339e-05, | |
| "loss": 0.7981, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.5062289895194779, | |
| "eval_loss": 0.6757128834724426, | |
| "eval_runtime": 85.0876, | |
| "eval_samples_per_second": 6.264, | |
| "eval_steps_per_second": 3.138, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5125568518884714, | |
| "grad_norm": 0.8669189810752869, | |
| "learning_rate": 8.047222744854943e-06, | |
| "loss": 0.685, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.5315404389954519, | |
| "grad_norm": 0.7252311110496521, | |
| "learning_rate": 3.6037139304146762e-06, | |
| "loss": 0.684, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.5505240261024322, | |
| "grad_norm": 0.9997105598449707, | |
| "learning_rate": 9.0502382320653e-07, | |
| "loss": 0.7458, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.5568518884714257, | |
| "eval_loss": 0.6707542538642883, | |
| "eval_runtime": 85.059, | |
| "eval_samples_per_second": 6.266, | |
| "eval_steps_per_second": 3.139, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.5695076132094127, | |
| "grad_norm": 0.8084313869476318, | |
| "learning_rate": 0.0, | |
| "loss": 0.6271, | |
| "step": 90 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 90, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.388734427267072e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |