| { | |
| "best_metric": 0.01498666312545538, | |
| "best_model_checkpoint": "./text2sql/codellama_instruct_pt_text2sql/checkpoint-44000", | |
| "epoch": 4.9995126366358, | |
| "eval_steps": 2000, | |
| "global_step": 44880, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1.1139992573338283e-07, | |
| "loss": 1.2191, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00011139992573338284, | |
| "loss": 0.2684, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00022279985146676567, | |
| "loss": 0.0693, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 0.05887996032834053, | |
| "eval_runtime": 171.127, | |
| "eval_samples_per_second": 11.687, | |
| "eval_steps_per_second": 1.461, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00029781686301467276, | |
| "loss": 0.0565, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0002907056676227274, | |
| "loss": 0.047, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_loss": 0.03957173973321915, | |
| "eval_runtime": 171.3682, | |
| "eval_samples_per_second": 11.671, | |
| "eval_steps_per_second": 1.459, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00028359447223078195, | |
| "loss": 0.0403, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002764832768388366, | |
| "loss": 0.0364, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 0.030724667012691498, | |
| "eval_runtime": 170.9405, | |
| "eval_samples_per_second": 11.7, | |
| "eval_steps_per_second": 1.462, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0002693720814468912, | |
| "loss": 0.0341, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00026226088605494583, | |
| "loss": 0.0311, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_loss": 0.027806701138615608, | |
| "eval_runtime": 171.064, | |
| "eval_samples_per_second": 11.692, | |
| "eval_steps_per_second": 1.461, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00025514969066300046, | |
| "loss": 0.0289, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.00024803849527105503, | |
| "loss": 0.0251, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_loss": 0.024082325398921967, | |
| "eval_runtime": 171.4689, | |
| "eval_samples_per_second": 11.664, | |
| "eval_steps_per_second": 1.458, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00024092729987910966, | |
| "loss": 0.0242, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.00023381610448716428, | |
| "loss": 0.0243, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "eval_loss": 0.02279273048043251, | |
| "eval_runtime": 171.2147, | |
| "eval_samples_per_second": 11.681, | |
| "eval_steps_per_second": 1.46, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00022670490909521888, | |
| "loss": 0.0219, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0002195937137032735, | |
| "loss": 0.0227, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_loss": 0.022252563387155533, | |
| "eval_runtime": 171.1834, | |
| "eval_samples_per_second": 11.683, | |
| "eval_steps_per_second": 1.46, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00021248251831132813, | |
| "loss": 0.0218, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.00020537132291938273, | |
| "loss": 0.0212, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_loss": 0.020056582987308502, | |
| "eval_runtime": 170.9191, | |
| "eval_samples_per_second": 11.701, | |
| "eval_steps_per_second": 1.463, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.00019826012752743733, | |
| "loss": 0.0208, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.00019114893213549196, | |
| "loss": 0.0202, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 0.01821763999760151, | |
| "eval_runtime": 171.4736, | |
| "eval_samples_per_second": 11.664, | |
| "eval_steps_per_second": 1.458, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 0.00018403773674354658, | |
| "loss": 0.0161, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.0001769265413516012, | |
| "loss": 0.016, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "eval_loss": 0.018352985382080078, | |
| "eval_runtime": 171.2111, | |
| "eval_samples_per_second": 11.681, | |
| "eval_steps_per_second": 1.46, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 0.00016981534595965578, | |
| "loss": 0.0155, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 0.0001627041505677104, | |
| "loss": 0.0156, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "eval_loss": 0.017868032678961754, | |
| "eval_runtime": 171.2047, | |
| "eval_samples_per_second": 11.682, | |
| "eval_steps_per_second": 1.46, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.00015559295517576503, | |
| "loss": 0.0155, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.00014848175978381966, | |
| "loss": 0.015, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_loss": 0.017283763736486435, | |
| "eval_runtime": 171.6023, | |
| "eval_samples_per_second": 11.655, | |
| "eval_steps_per_second": 1.457, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 0.00014137056439187426, | |
| "loss": 0.0153, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 0.00013425936899992888, | |
| "loss": 0.0147, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_loss": 0.016522083431482315, | |
| "eval_runtime": 171.211, | |
| "eval_samples_per_second": 11.681, | |
| "eval_steps_per_second": 1.46, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 0.00012714817360798348, | |
| "loss": 0.0137, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.00012003697821603811, | |
| "loss": 0.0112, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_loss": 0.016529470682144165, | |
| "eval_runtime": 170.9168, | |
| "eval_samples_per_second": 11.702, | |
| "eval_steps_per_second": 1.463, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.00011292578282409272, | |
| "loss": 0.0112, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 0.00010581458743214733, | |
| "loss": 0.0109, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "eval_loss": 0.016144245862960815, | |
| "eval_runtime": 171.5226, | |
| "eval_samples_per_second": 11.66, | |
| "eval_steps_per_second": 1.458, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 9.870339204020194e-05, | |
| "loss": 0.0111, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 9.159219664825657e-05, | |
| "loss": 0.0109, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "eval_loss": 0.015489176847040653, | |
| "eval_runtime": 171.4873, | |
| "eval_samples_per_second": 11.663, | |
| "eval_steps_per_second": 1.458, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 8.448100125631117e-05, | |
| "loss": 0.0111, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 7.73698058643658e-05, | |
| "loss": 0.0105, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "eval_loss": 0.015172351151704788, | |
| "eval_runtime": 171.4626, | |
| "eval_samples_per_second": 11.664, | |
| "eval_steps_per_second": 1.458, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 7.025861047242041e-05, | |
| "loss": 0.0107, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 6.314741508047502e-05, | |
| "loss": 0.0104, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "eval_loss": 0.014995447359979153, | |
| "eval_runtime": 171.1146, | |
| "eval_samples_per_second": 11.688, | |
| "eval_steps_per_second": 1.461, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 5.603621968852963e-05, | |
| "loss": 0.0081, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 4.8925024296584245e-05, | |
| "loss": 0.0077, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "eval_loss": 0.015779005363583565, | |
| "eval_runtime": 171.0935, | |
| "eval_samples_per_second": 11.69, | |
| "eval_steps_per_second": 1.461, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 4.181382890463887e-05, | |
| "loss": 0.0079, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 3.4702633512693484e-05, | |
| "loss": 0.0078, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "eval_loss": 0.015086671337485313, | |
| "eval_runtime": 171.2287, | |
| "eval_samples_per_second": 11.68, | |
| "eval_steps_per_second": 1.46, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 2.7591438120748093e-05, | |
| "loss": 0.0079, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 2.0480242728802712e-05, | |
| "loss": 0.0076, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "eval_loss": 0.015026736073195934, | |
| "eval_runtime": 170.9263, | |
| "eval_samples_per_second": 11.701, | |
| "eval_steps_per_second": 1.463, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 1.3369047336857326e-05, | |
| "loss": 0.0075, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 6.257851944911939e-06, | |
| "loss": 0.0077, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_loss": 0.01498666312545538, | |
| "eval_runtime": 171.2767, | |
| "eval_samples_per_second": 11.677, | |
| "eval_steps_per_second": 1.46, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 44880, | |
| "total_flos": 2.184131254639421e+19, | |
| "train_loss": 0.0249389049336447, | |
| "train_runtime": 262166.9813, | |
| "train_samples_per_second": 5.479, | |
| "train_steps_per_second": 0.171 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 44880, | |
| "num_train_epochs": 5, | |
| "save_steps": 2000, | |
| "total_flos": 2.184131254639421e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |