| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 1000, | |
| "global_step": 11346, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13220518244315177, | |
| "grad_norm": 0.8546377420425415, | |
| "learning_rate": 8.816009873931059e-05, | |
| "loss": 5.1118, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.26441036488630354, | |
| "grad_norm": 0.8593683838844299, | |
| "learning_rate": 9.59831475011252e-05, | |
| "loss": 3.406, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.26441036488630354, | |
| "eval_accuracy": 0.5035308829464115, | |
| "eval_loss": 3.23445987701416, | |
| "eval_runtime": 128.969, | |
| "eval_samples_per_second": 14.267, | |
| "eval_steps_per_second": 0.302, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3966155473294553, | |
| "grad_norm": 0.9617242217063904, | |
| "learning_rate": 9.134314230431938e-05, | |
| "loss": 3.0005, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.5288207297726071, | |
| "grad_norm": 0.8956136107444763, | |
| "learning_rate": 8.670313710751356e-05, | |
| "loss": 2.8119, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.5288207297726071, | |
| "eval_accuracy": 0.5364783564495231, | |
| "eval_loss": 2.821624517440796, | |
| "eval_runtime": 128.8547, | |
| "eval_samples_per_second": 14.28, | |
| "eval_steps_per_second": 0.303, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6610259122157589, | |
| "grad_norm": 1.2592207193374634, | |
| "learning_rate": 8.206313191070773e-05, | |
| "loss": 2.6861, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.7932310946589106, | |
| "grad_norm": 1.3978203535079956, | |
| "learning_rate": 7.742312671390191e-05, | |
| "loss": 2.6076, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.7932310946589106, | |
| "eval_accuracy": 0.5501396529385527, | |
| "eval_loss": 2.6552908420562744, | |
| "eval_runtime": 129.1173, | |
| "eval_samples_per_second": 14.251, | |
| "eval_steps_per_second": 0.302, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.9254362771020624, | |
| "grad_norm": 1.8927521705627441, | |
| "learning_rate": 7.278312151709609e-05, | |
| "loss": 2.5643, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.0576414595452142, | |
| "grad_norm": 4.235791206359863, | |
| "learning_rate": 6.814311632029027e-05, | |
| "loss": 2.4729, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.0576414595452142, | |
| "eval_accuracy": 0.5581473949151462, | |
| "eval_loss": 2.5761468410491943, | |
| "eval_runtime": 129.3204, | |
| "eval_samples_per_second": 14.228, | |
| "eval_steps_per_second": 0.302, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.189846641988366, | |
| "grad_norm": 2.753159523010254, | |
| "learning_rate": 6.350311112348446e-05, | |
| "loss": 2.4531, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.3220518244315178, | |
| "grad_norm": 3.542167901992798, | |
| "learning_rate": 5.886310592667864e-05, | |
| "loss": 2.4323, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.3220518244315178, | |
| "eval_accuracy": 0.5616867738578195, | |
| "eval_loss": 2.536275863647461, | |
| "eval_runtime": 129.4842, | |
| "eval_samples_per_second": 14.21, | |
| "eval_steps_per_second": 0.301, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.4542570068746694, | |
| "grad_norm": 3.678074598312378, | |
| "learning_rate": 5.422310072987282e-05, | |
| "loss": 2.3981, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.5864621893178212, | |
| "grad_norm": 5.024658679962158, | |
| "learning_rate": 4.9583095533066995e-05, | |
| "loss": 2.3824, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.5864621893178212, | |
| "eval_accuracy": 0.5660391134427901, | |
| "eval_loss": 2.491274118423462, | |
| "eval_runtime": 129.1418, | |
| "eval_samples_per_second": 14.248, | |
| "eval_steps_per_second": 0.302, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.718667371760973, | |
| "grad_norm": 5.106564521789551, | |
| "learning_rate": 4.4943090336261176e-05, | |
| "loss": 2.3763, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.8508725542041247, | |
| "grad_norm": 11.649778366088867, | |
| "learning_rate": 4.030308513945535e-05, | |
| "loss": 2.3719, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.8508725542041247, | |
| "eval_accuracy": 0.5685828678235382, | |
| "eval_loss": 2.466362237930298, | |
| "eval_runtime": 128.8316, | |
| "eval_samples_per_second": 14.282, | |
| "eval_steps_per_second": 0.303, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.9830777366472767, | |
| "grad_norm": 5.842829704284668, | |
| "learning_rate": 3.566307994264953e-05, | |
| "loss": 2.3411, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.1152829190904283, | |
| "grad_norm": 14.795243263244629, | |
| "learning_rate": 3.1023074745843715e-05, | |
| "loss": 2.3021, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.1152829190904283, | |
| "eval_accuracy": 0.5715702725090801, | |
| "eval_loss": 2.4403789043426514, | |
| "eval_runtime": 128.3806, | |
| "eval_samples_per_second": 14.332, | |
| "eval_steps_per_second": 0.304, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.24748810153358, | |
| "grad_norm": 6.084632873535156, | |
| "learning_rate": 2.6383069549037897e-05, | |
| "loss": 2.2897, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.379693283976732, | |
| "grad_norm": 17.67453956604004, | |
| "learning_rate": 2.1743064352232075e-05, | |
| "loss": 2.2848, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.379693283976732, | |
| "eval_accuracy": 0.5754845372868036, | |
| "eval_loss": 2.407994508743286, | |
| "eval_runtime": 128.557, | |
| "eval_samples_per_second": 14.313, | |
| "eval_steps_per_second": 0.303, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.5118984664198836, | |
| "grad_norm": 6.01920223236084, | |
| "learning_rate": 1.7103059155426253e-05, | |
| "loss": 2.2611, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.6441036488630356, | |
| "grad_norm": 7.932415008544922, | |
| "learning_rate": 1.2463053958620433e-05, | |
| "loss": 2.2653, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.6441036488630356, | |
| "eval_accuracy": 0.5784841549669718, | |
| "eval_loss": 2.3834304809570312, | |
| "eval_runtime": 128.7742, | |
| "eval_samples_per_second": 14.289, | |
| "eval_steps_per_second": 0.303, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.776308831306187, | |
| "grad_norm": 9.934358596801758, | |
| "learning_rate": 7.823048761814613e-06, | |
| "loss": 2.2542, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.908514013749339, | |
| "grad_norm": 8.533864974975586, | |
| "learning_rate": 3.183043565008793e-06, | |
| "loss": 2.2447, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.908514013749339, | |
| "eval_accuracy": 0.5811017183152439, | |
| "eval_loss": 2.3603451251983643, | |
| "eval_runtime": 129.0604, | |
| "eval_samples_per_second": 14.257, | |
| "eval_steps_per_second": 0.302, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 11346, | |
| "total_flos": 1.4536404559724544e+17, | |
| "train_loss": 2.5941595100713495, | |
| "train_runtime": 20556.3593, | |
| "train_samples_per_second": 13.243, | |
| "train_steps_per_second": 0.552 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 11346, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4536404559724544e+17, | |
| "train_batch_size": 24, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |