| { | |
| "best_metric": 0.01576746255159378, | |
| "best_model_checkpoint": "./text2sql/13b-codellama_instruct_pt_text2sql/checkpoint-36000", | |
| "epoch": 4.233128407215813, | |
| "eval_steps": 2000, | |
| "global_step": 38000, | |
| "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": 1.0897, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00011139992573338284, | |
| "loss": 0.2546, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00022279985146676567, | |
| "loss": 0.0774, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 0.0608125701546669, | |
| "eval_runtime": 285.1249, | |
| "eval_samples_per_second": 7.014, | |
| "eval_steps_per_second": 1.754, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00029781686301467276, | |
| "loss": 0.0652, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.0002907056676227274, | |
| "loss": 0.0581, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_loss": 0.04370184242725372, | |
| "eval_runtime": 284.9797, | |
| "eval_samples_per_second": 7.018, | |
| "eval_steps_per_second": 1.755, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00028359447223078195, | |
| "loss": 0.0499, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0002764832768388366, | |
| "loss": 0.0445, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 0.036812592297792435, | |
| "eval_runtime": 284.5683, | |
| "eval_samples_per_second": 7.028, | |
| "eval_steps_per_second": 1.757, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0002693720814468912, | |
| "loss": 0.0433, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00026226088605494583, | |
| "loss": 0.0378, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_loss": 0.03097330778837204, | |
| "eval_runtime": 354.6673, | |
| "eval_samples_per_second": 5.639, | |
| "eval_steps_per_second": 1.41, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00025514969066300046, | |
| "loss": 0.0364, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.00024803849527105503, | |
| "loss": 0.0312, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_loss": 0.02841157279908657, | |
| "eval_runtime": 355.7754, | |
| "eval_samples_per_second": 5.622, | |
| "eval_steps_per_second": 1.405, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00024092729987910966, | |
| "loss": 0.0307, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.00023381610448716428, | |
| "loss": 0.0316, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "eval_loss": 0.02577413246035576, | |
| "eval_runtime": 284.3906, | |
| "eval_samples_per_second": 7.033, | |
| "eval_steps_per_second": 1.758, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00022670490909521888, | |
| "loss": 0.0272, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0002195937137032735, | |
| "loss": 0.0279, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_loss": 0.024905819445848465, | |
| "eval_runtime": 285.1649, | |
| "eval_samples_per_second": 7.013, | |
| "eval_steps_per_second": 1.753, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00021248251831132813, | |
| "loss": 0.0285, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.00020537132291938273, | |
| "loss": 0.0266, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_loss": 0.023692620918154716, | |
| "eval_runtime": 285.1334, | |
| "eval_samples_per_second": 7.014, | |
| "eval_steps_per_second": 1.754, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.00019826012752743733, | |
| "loss": 0.0266, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.00019114893213549196, | |
| "loss": 0.0258, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 0.021677831187844276, | |
| "eval_runtime": 285.282, | |
| "eval_samples_per_second": 7.011, | |
| "eval_steps_per_second": 1.753, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 0.00018403773674354658, | |
| "loss": 0.0209, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.0001769265413516012, | |
| "loss": 0.0211, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "eval_loss": 0.020885193720459938, | |
| "eval_runtime": 284.9168, | |
| "eval_samples_per_second": 7.02, | |
| "eval_steps_per_second": 1.755, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 0.00016981534595965578, | |
| "loss": 0.0209, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 0.0001627041505677104, | |
| "loss": 0.0194, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "eval_loss": 0.020141348242759705, | |
| "eval_runtime": 285.3239, | |
| "eval_samples_per_second": 7.01, | |
| "eval_steps_per_second": 1.752, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 0.00015559295517576503, | |
| "loss": 0.0212, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.00014848175978381966, | |
| "loss": 0.0202, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_loss": 0.019859744235873222, | |
| "eval_runtime": 285.2272, | |
| "eval_samples_per_second": 7.012, | |
| "eval_steps_per_second": 1.753, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 0.00014137056439187426, | |
| "loss": 0.0191, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 0.00013425936899992888, | |
| "loss": 0.0183, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_loss": 0.01871539279818535, | |
| "eval_runtime": 285.1023, | |
| "eval_samples_per_second": 7.015, | |
| "eval_steps_per_second": 1.754, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 0.00012714817360798348, | |
| "loss": 0.0175, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.00012003697821603811, | |
| "loss": 0.0158, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_loss": 0.017088208347558975, | |
| "eval_runtime": 284.6987, | |
| "eval_samples_per_second": 7.025, | |
| "eval_steps_per_second": 1.756, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.00011292578282409272, | |
| "loss": 0.0154, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 0.00010581458743214733, | |
| "loss": 0.0143, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "eval_loss": 0.017135081812739372, | |
| "eval_runtime": 284.2489, | |
| "eval_samples_per_second": 7.036, | |
| "eval_steps_per_second": 1.759, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 9.870339204020194e-05, | |
| "loss": 0.015, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 9.159219664825657e-05, | |
| "loss": 0.0148, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "eval_loss": 0.017010025680065155, | |
| "eval_runtime": 284.4395, | |
| "eval_samples_per_second": 7.031, | |
| "eval_steps_per_second": 1.758, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 8.448100125631117e-05, | |
| "loss": 0.0148, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 7.73698058643658e-05, | |
| "loss": 0.013, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "eval_loss": 0.01629829593002796, | |
| "eval_runtime": 284.3551, | |
| "eval_samples_per_second": 7.033, | |
| "eval_steps_per_second": 1.758, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 7.025861047242041e-05, | |
| "loss": 0.0139, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 6.314741508047502e-05, | |
| "loss": 0.0129, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "eval_loss": 0.01576746255159378, | |
| "eval_runtime": 284.6218, | |
| "eval_samples_per_second": 7.027, | |
| "eval_steps_per_second": 1.757, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 5.603621968852963e-05, | |
| "loss": 0.0104, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 4.8925024296584245e-05, | |
| "loss": 0.0104, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "eval_loss": 0.016835128888487816, | |
| "eval_runtime": 284.6011, | |
| "eval_samples_per_second": 7.027, | |
| "eval_steps_per_second": 1.757, | |
| "step": 38000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 44880, | |
| "num_train_epochs": 5, | |
| "save_steps": 2000, | |
| "total_flos": 2.617032039948288e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |