| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 8240, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12135922330097088, | |
| "grad_norm": 0.9599943161010742, | |
| "learning_rate": 9.698956057295461e-05, | |
| "loss": 1.2196, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.24271844660194175, | |
| "grad_norm": 1.0285232067108154, | |
| "learning_rate": 9.395484340859432e-05, | |
| "loss": 0.6688, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3640776699029126, | |
| "grad_norm": 0.8658091425895691, | |
| "learning_rate": 9.092012624423404e-05, | |
| "loss": 0.5603, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.4854368932038835, | |
| "grad_norm": 0.8770154714584351, | |
| "learning_rate": 8.788540907987377e-05, | |
| "loss": 0.5165, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6067961165048543, | |
| "grad_norm": 0.6812583208084106, | |
| "learning_rate": 8.485069191551348e-05, | |
| "loss": 0.5015, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.7281553398058253, | |
| "grad_norm": 0.6911689043045044, | |
| "learning_rate": 8.181597475115321e-05, | |
| "loss": 0.4846, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8495145631067961, | |
| "grad_norm": 0.6556753516197205, | |
| "learning_rate": 7.878125758679291e-05, | |
| "loss": 0.4769, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.970873786407767, | |
| "grad_norm": 0.5876182317733765, | |
| "learning_rate": 7.574654042243264e-05, | |
| "loss": 0.4701, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0922330097087378, | |
| "grad_norm": 0.6175569891929626, | |
| "learning_rate": 7.271182325807235e-05, | |
| "loss": 0.4616, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.2135922330097086, | |
| "grad_norm": 0.6353004574775696, | |
| "learning_rate": 6.967710609371208e-05, | |
| "loss": 0.4518, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.3349514563106797, | |
| "grad_norm": 0.5879459977149963, | |
| "learning_rate": 6.664238892935178e-05, | |
| "loss": 0.4483, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 1.4563106796116505, | |
| "grad_norm": 0.6575189232826233, | |
| "learning_rate": 6.360767176499151e-05, | |
| "loss": 0.4367, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.5776699029126213, | |
| "grad_norm": 0.724533200263977, | |
| "learning_rate": 6.0572954600631224e-05, | |
| "loss": 0.4226, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 1.6990291262135924, | |
| "grad_norm": 0.7686433792114258, | |
| "learning_rate": 5.7538237436270945e-05, | |
| "loss": 0.4104, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.820388349514563, | |
| "grad_norm": 0.7101556658744812, | |
| "learning_rate": 5.450352027191066e-05, | |
| "loss": 0.3954, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 1.941747572815534, | |
| "grad_norm": 0.7856088280677795, | |
| "learning_rate": 5.146880310755038e-05, | |
| "loss": 0.3827, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.063106796116505, | |
| "grad_norm": 0.8785816431045532, | |
| "learning_rate": 4.84340859431901e-05, | |
| "loss": 0.3585, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 2.1844660194174756, | |
| "grad_norm": 0.858726441860199, | |
| "learning_rate": 4.539936877882982e-05, | |
| "loss": 0.341, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.3058252427184467, | |
| "grad_norm": 0.8789017200469971, | |
| "learning_rate": 4.236465161446954e-05, | |
| "loss": 0.3313, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 2.4271844660194173, | |
| "grad_norm": 0.9984813928604126, | |
| "learning_rate": 3.932993445010925e-05, | |
| "loss": 0.321, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.5485436893203883, | |
| "grad_norm": 0.8649771213531494, | |
| "learning_rate": 3.6295217285748975e-05, | |
| "loss": 0.3104, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 2.6699029126213594, | |
| "grad_norm": 0.9905620217323303, | |
| "learning_rate": 3.326050012138869e-05, | |
| "loss": 0.3008, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.79126213592233, | |
| "grad_norm": 0.9460727572441101, | |
| "learning_rate": 3.022578295702841e-05, | |
| "loss": 0.2965, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 2.912621359223301, | |
| "grad_norm": 0.8885589241981506, | |
| "learning_rate": 2.7191065792668125e-05, | |
| "loss": 0.2876, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.033980582524272, | |
| "grad_norm": 0.9261214733123779, | |
| "learning_rate": 2.4156348628307843e-05, | |
| "loss": 0.2759, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 3.1553398058252426, | |
| "grad_norm": 0.9241772294044495, | |
| "learning_rate": 2.112163146394756e-05, | |
| "loss": 0.2618, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.2766990291262137, | |
| "grad_norm": 0.929602861404419, | |
| "learning_rate": 1.808691429958728e-05, | |
| "loss": 0.2578, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 3.3980582524271843, | |
| "grad_norm": 0.9885833263397217, | |
| "learning_rate": 1.5052197135226997e-05, | |
| "loss": 0.2547, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.5194174757281553, | |
| "grad_norm": 0.9474493861198425, | |
| "learning_rate": 1.2017479970866715e-05, | |
| "loss": 0.2528, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 3.6407766990291264, | |
| "grad_norm": 0.9105657935142517, | |
| "learning_rate": 8.982762806506435e-06, | |
| "loss": 0.25, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.762135922330097, | |
| "grad_norm": 0.9185407161712646, | |
| "learning_rate": 5.948045642146152e-06, | |
| "loss": 0.2463, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 3.883495145631068, | |
| "grad_norm": 0.8763870000839233, | |
| "learning_rate": 2.9133284777858704e-06, | |
| "loss": 0.2462, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 8240, | |
| "total_flos": 1.1079720316327956e+18, | |
| "train_loss": 0.39851856185394585, | |
| "train_runtime": 97554.2615, | |
| "train_samples_per_second": 5.406, | |
| "train_steps_per_second": 0.084 | |
| } | |
| ], | |
| "logging_steps": 250, | |
| "max_steps": 8240, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1079720316327956e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |