| { | |
| "best_metric": 0.35073795914649963, | |
| "best_model_checkpoint": "../../saves/LLaMA3-70B-qlora-bnb/lora/sft/A61K-15950_2/checkpoint-200", | |
| "epoch": 0.9975308641975309, | |
| "eval_steps": 100, | |
| "global_step": 202, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04938271604938271, | |
| "grad_norm": 46.31583786010742, | |
| "learning_rate": 9.523809523809523e-06, | |
| "loss": 12.5496, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.09876543209876543, | |
| "grad_norm": 23.47727394104004, | |
| "learning_rate": 5.714285714285714e-05, | |
| "loss": 11.9318, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.14814814814814814, | |
| "grad_norm": 25.78252410888672, | |
| "learning_rate": 9.999246866958692e-05, | |
| "loss": 7.8505, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 13.642712593078613, | |
| "learning_rate": 9.909145108993794e-05, | |
| "loss": 1.2163, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.24691358024691357, | |
| "grad_norm": 10.011795997619629, | |
| "learning_rate": 9.67152097716334e-05, | |
| "loss": 0.4577, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2962962962962963, | |
| "grad_norm": 10.739720344543457, | |
| "learning_rate": 9.29351520070574e-05, | |
| "loss": 0.3768, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.345679012345679, | |
| "grad_norm": 11.204980850219727, | |
| "learning_rate": 8.786487050581583e-05, | |
| "loss": 0.4593, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "grad_norm": 4.646059513092041, | |
| "learning_rate": 8.165672987449962e-05, | |
| "loss": 0.4234, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 5.678186893463135, | |
| "learning_rate": 7.449728798069864e-05, | |
| "loss": 0.3857, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.49382716049382713, | |
| "grad_norm": 49.75868225097656, | |
| "learning_rate": 6.66016897916682e-05, | |
| "loss": 0.3663, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.49382716049382713, | |
| "eval_loss": 0.35489073395729065, | |
| "eval_runtime": 1317.5223, | |
| "eval_samples_per_second": 0.137, | |
| "eval_steps_per_second": 0.137, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5432098765432098, | |
| "grad_norm": 2.499194860458374, | |
| "learning_rate": 5.820720215572375e-05, | |
| "loss": 0.37, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 3.846980571746826, | |
| "learning_rate": 4.956608380955877e-05, | |
| "loss": 0.3788, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.6419753086419753, | |
| "grad_norm": 2.995356321334839, | |
| "learning_rate": 4.093800487148857e-05, | |
| "loss": 0.401, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.691358024691358, | |
| "grad_norm": 4.932299613952637, | |
| "learning_rate": 3.258224361880657e-05, | |
| "loss": 0.3944, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.7407407407407407, | |
| "grad_norm": 4.609350681304932, | |
| "learning_rate": 2.474989504016798e-05, | |
| "loss": 0.3521, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "grad_norm": 3.664435625076294, | |
| "learning_rate": 1.7676325300069825e-05, | |
| "loss": 0.3425, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.8395061728395061, | |
| "grad_norm": 31.691667556762695, | |
| "learning_rate": 1.1574098862709992e-05, | |
| "loss": 0.3548, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 2.024195909500122, | |
| "learning_rate": 6.626590818846162e-06, | |
| "loss": 0.3404, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.9382716049382716, | |
| "grad_norm": 2.383239507675171, | |
| "learning_rate": 2.9824763685681766e-06, | |
| "loss": 0.3374, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.9876543209876543, | |
| "grad_norm": 2.007232189178467, | |
| "learning_rate": 7.512630539036502e-07, | |
| "loss": 0.3383, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9876543209876543, | |
| "eval_loss": 0.35073795914649963, | |
| "eval_runtime": 1316.6761, | |
| "eval_samples_per_second": 0.137, | |
| "eval_steps_per_second": 0.137, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9975308641975309, | |
| "step": 202, | |
| "total_flos": 8.350412130662744e+18, | |
| "train_loss": 1.9651573648547183, | |
| "train_runtime": 27220.4722, | |
| "train_samples_per_second": 0.06, | |
| "train_steps_per_second": 0.007 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 202, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 8.350412130662744e+18, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |