| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.9912280701754386, |
| "eval_steps": 500, |
| "global_step": 113, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.043859649122807015, |
| "grad_norm": 41.67070770263672, |
| "learning_rate": 4.081632653061224e-06, |
| "loss": 1.5701, |
| "mean_token_accuracy": 0.8569930195808411, |
| "num_tokens": 513024.0, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.08771929824561403, |
| "grad_norm": 2.8569185733795166, |
| "learning_rate": 9.183673469387754e-06, |
| "loss": 0.5994, |
| "mean_token_accuracy": 0.8944263458251953, |
| "num_tokens": 1038096.0, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.13157894736842105, |
| "grad_norm": 1.5813679695129395, |
| "learning_rate": 1.4285714285714285e-05, |
| "loss": 0.2119, |
| "mean_token_accuracy": 0.925440788269043, |
| "num_tokens": 1548448.0, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.17543859649122806, |
| "grad_norm": 1.0589200258255005, |
| "learning_rate": 1.9387755102040814e-05, |
| "loss": 0.1818, |
| "mean_token_accuracy": 0.9331316351890564, |
| "num_tokens": 2073817.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.21929824561403508, |
| "grad_norm": 1.3676564693450928, |
| "learning_rate": 2.4489795918367345e-05, |
| "loss": 0.1406, |
| "mean_token_accuracy": 0.9494450569152832, |
| "num_tokens": 2608742.0, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.2631578947368421, |
| "grad_norm": 1.3096325397491455, |
| "learning_rate": 2.9591836734693876e-05, |
| "loss": 0.1081, |
| "mean_token_accuracy": 0.9615433573722839, |
| "num_tokens": 3133080.0, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.30701754385964913, |
| "grad_norm": 1.0308455228805542, |
| "learning_rate": 3.4693877551020405e-05, |
| "loss": 0.0956, |
| "mean_token_accuracy": 0.9669486165046692, |
| "num_tokens": 3651893.0, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.3508771929824561, |
| "grad_norm": 0.9645959734916687, |
| "learning_rate": 3.5545205472361604e-05, |
| "loss": 0.0783, |
| "mean_token_accuracy": 0.973760974407196, |
| "num_tokens": 4172274.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.39473684210526316, |
| "grad_norm": 0.9264997839927673, |
| "learning_rate": 3.486562720895617e-05, |
| "loss": 0.0691, |
| "mean_token_accuracy": 0.9768692851066589, |
| "num_tokens": 4679648.0, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.43859649122807015, |
| "grad_norm": 1.101047396659851, |
| "learning_rate": 3.3691741456580764e-05, |
| "loss": 0.0762, |
| "mean_token_accuracy": 0.972927701473236, |
| "num_tokens": 5214708.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.4824561403508772, |
| "grad_norm": 0.8863373398780823, |
| "learning_rate": 3.206980549602e-05, |
| "loss": 0.0655, |
| "mean_token_accuracy": 0.9797237038612365, |
| "num_tokens": 5756390.0, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.5263157894736842, |
| "grad_norm": 0.8536093831062317, |
| "learning_rate": 3.0063732145868502e-05, |
| "loss": 0.0599, |
| "mean_token_accuracy": 0.9814828395843506, |
| "num_tokens": 6291718.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.5701754385964912, |
| "grad_norm": 1.0385499000549316, |
| "learning_rate": 2.7752571260888306e-05, |
| "loss": 0.0613, |
| "mean_token_accuracy": 0.9785101056098938, |
| "num_tokens": 6819802.0, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.6140350877192983, |
| "grad_norm": 0.8211565017700195, |
| "learning_rate": 2.52273947514239e-05, |
| "loss": 0.0529, |
| "mean_token_accuracy": 0.982378077507019, |
| "num_tokens": 7364188.0, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.6578947368421053, |
| "grad_norm": 0.747939944267273, |
| "learning_rate": 2.258770787051433e-05, |
| "loss": 0.0526, |
| "mean_token_accuracy": 0.9829726338386535, |
| "num_tokens": 7893505.0, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.7017543859649122, |
| "grad_norm": 0.7447258830070496, |
| "learning_rate": 1.993752818288995e-05, |
| "loss": 0.0448, |
| "mean_token_accuracy": 0.9852114915847778, |
| "num_tokens": 8396944.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.7456140350877193, |
| "grad_norm": 0.483532577753067, |
| "learning_rate": 1.7381286725124893e-05, |
| "loss": 0.0451, |
| "mean_token_accuracy": 0.9849389910697937, |
| "num_tokens": 8936137.0, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.7894736842105263, |
| "grad_norm": 0.5197746157646179, |
| "learning_rate": 1.5019712872821418e-05, |
| "loss": 0.0501, |
| "mean_token_accuracy": 0.9841407418251038, |
| "num_tokens": 9478103.0, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.8333333333333334, |
| "grad_norm": 0.6424993276596069, |
| "learning_rate": 1.294586507273091e-05, |
| "loss": 0.0408, |
| "mean_token_accuracy": 0.9856717944145202, |
| "num_tokens": 10012762.0, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.8771929824561403, |
| "grad_norm": 0.5072142481803894, |
| "learning_rate": 1.124146384986977e-05, |
| "loss": 0.0433, |
| "mean_token_accuracy": 0.9873993396759033, |
| "num_tokens": 10547128.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.9210526315789473, |
| "grad_norm": 0.8326196074485779, |
| "learning_rate": 9.973671588461518e-06, |
| "loss": 0.0432, |
| "mean_token_accuracy": 0.9862016916275025, |
| "num_tokens": 11066399.0, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.9649122807017544, |
| "grad_norm": 0.7005944848060608, |
| "learning_rate": 9.192445980295052e-06, |
| "loss": 0.0415, |
| "mean_token_accuracy": 0.9858503937721252, |
| "num_tokens": 11594955.0, |
| "step": 110 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 114, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.7847980825234637e+17, |
| "train_batch_size": 20, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
