| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0009620183144236609, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.810091572118304e-05, | |
| "grad_norm": 1.8092963695526123, | |
| "learning_rate": 1e-05, | |
| "loss": 4.5577, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 4.810091572118304e-05, | |
| "eval_loss": 4.658426761627197, | |
| "eval_runtime": 938.5736, | |
| "eval_samples_per_second": 37.307, | |
| "eval_steps_per_second": 18.654, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 9.620183144236608e-05, | |
| "grad_norm": 6.848754405975342, | |
| "learning_rate": 2e-05, | |
| "loss": 4.5374, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00014430274716354912, | |
| "grad_norm": 1.8597630262374878, | |
| "learning_rate": 3e-05, | |
| "loss": 4.2371, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00019240366288473217, | |
| "grad_norm": 1.916515588760376, | |
| "learning_rate": 4e-05, | |
| "loss": 4.5267, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00024050457860591521, | |
| "grad_norm": 1.4189711809158325, | |
| "learning_rate": 5e-05, | |
| "loss": 4.1488, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00024050457860591521, | |
| "eval_loss": 4.593773365020752, | |
| "eval_runtime": 842.331, | |
| "eval_samples_per_second": 41.569, | |
| "eval_steps_per_second": 20.785, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00028860549432709824, | |
| "grad_norm": 1.630517601966858, | |
| "learning_rate": 6e-05, | |
| "loss": 4.1124, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0003367064100482813, | |
| "grad_norm": 2.1663177013397217, | |
| "learning_rate": 7e-05, | |
| "loss": 4.2538, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00038480732576946433, | |
| "grad_norm": 1.7669731378555298, | |
| "learning_rate": 8e-05, | |
| "loss": 4.4313, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0004329082414906474, | |
| "grad_norm": 2.2088398933410645, | |
| "learning_rate": 9e-05, | |
| "loss": 4.4259, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00048100915721183043, | |
| "grad_norm": 1.989790439605713, | |
| "learning_rate": 0.0001, | |
| "loss": 3.7162, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00048100915721183043, | |
| "eval_loss": 4.230820655822754, | |
| "eval_runtime": 843.7217, | |
| "eval_samples_per_second": 41.501, | |
| "eval_steps_per_second": 20.751, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0005291100729330134, | |
| "grad_norm": 2.0854384899139404, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 4.0409, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0005772109886541965, | |
| "grad_norm": 2.4684157371520996, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 3.6082, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0006253119043753795, | |
| "grad_norm": 2.6682002544403076, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 3.9602, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0006734128200965626, | |
| "grad_norm": 3.6990573406219482, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 3.4611, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0007215137358177456, | |
| "grad_norm": 12.550111770629883, | |
| "learning_rate": 5e-05, | |
| "loss": 3.8372, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0007215137358177456, | |
| "eval_loss": 3.7095682621002197, | |
| "eval_runtime": 843.6594, | |
| "eval_samples_per_second": 41.504, | |
| "eval_steps_per_second": 20.752, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0007696146515389287, | |
| "grad_norm": 1.8937203884124756, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 3.4878, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0008177155672601117, | |
| "grad_norm": 2.0600931644439697, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 3.581, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0008658164829812948, | |
| "grad_norm": 1.9563463926315308, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 3.4161, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0009139173987024778, | |
| "grad_norm": 2.333648204803467, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 3.2874, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0009620183144236609, | |
| "grad_norm": 2.211129903793335, | |
| "learning_rate": 0.0, | |
| "loss": 3.4981, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0009620183144236609, | |
| "eval_loss": 3.6328890323638916, | |
| "eval_runtime": 1003.4934, | |
| "eval_samples_per_second": 34.893, | |
| "eval_steps_per_second": 17.447, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.479651398516736e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |