| { | |
| "best_metric": 0.6521381139755249, | |
| "best_model_checkpoint": "/hy-tmp/checkpoints/zhongjing_7-13/checkpoint-6000", | |
| "epoch": 2.271221728021198, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 5.944240397524184e-05, | |
| "loss": 0.7616, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 0.7090576887130737, | |
| "eval_runtime": 1700.4285, | |
| "eval_samples_per_second": 20.932, | |
| "eval_steps_per_second": 0.328, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 5.7731724820150744e-05, | |
| "loss": 0.6972, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 0.6910683512687683, | |
| "eval_runtime": 1700.3181, | |
| "eval_samples_per_second": 20.933, | |
| "eval_steps_per_second": 0.328, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 5.4934584214999246e-05, | |
| "loss": 0.6845, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 0.6805794835090637, | |
| "eval_runtime": 1698.9819, | |
| "eval_samples_per_second": 20.95, | |
| "eval_steps_per_second": 0.328, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 5.116056731749404e-05, | |
| "loss": 0.6764, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 0.6737232804298401, | |
| "eval_runtime": 1702.5596, | |
| "eval_samples_per_second": 20.906, | |
| "eval_steps_per_second": 0.327, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 4.6557530919724635e-05, | |
| "loss": 0.6701, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 0.6684596538543701, | |
| "eval_runtime": 1702.1448, | |
| "eval_samples_per_second": 20.911, | |
| "eval_steps_per_second": 0.327, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 4.131682830461389e-05, | |
| "loss": 0.663, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_loss": 0.6644229292869568, | |
| "eval_runtime": 1699.1079, | |
| "eval_samples_per_second": 20.948, | |
| "eval_steps_per_second": 0.328, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3.5622841602995877e-05, | |
| "loss": 0.6594, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_loss": 0.661072313785553, | |
| "eval_runtime": 1696.6207, | |
| "eval_samples_per_second": 20.979, | |
| "eval_steps_per_second": 0.328, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 2.9720460586930557e-05, | |
| "loss": 0.6561, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_loss": 0.6583240032196045, | |
| "eval_runtime": 1699.0104, | |
| "eval_samples_per_second": 20.949, | |
| "eval_steps_per_second": 0.328, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 2.381734702787557e-05, | |
| "loss": 0.6538, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_loss": 0.6560451984405518, | |
| "eval_runtime": 1701.2915, | |
| "eval_samples_per_second": 20.921, | |
| "eval_steps_per_second": 0.327, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 1.815645475714211e-05, | |
| "loss": 0.6525, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 0.654275119304657, | |
| "eval_runtime": 1698.3722, | |
| "eval_samples_per_second": 20.957, | |
| "eval_steps_per_second": 0.328, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.2969355277673462e-05, | |
| "loss": 0.6488, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "eval_loss": 0.6530821919441223, | |
| "eval_runtime": 1698.6748, | |
| "eval_samples_per_second": 20.953, | |
| "eval_steps_per_second": 0.328, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 8.438544816619625e-06, | |
| "loss": 0.6478, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "eval_loss": 0.6521381139755249, | |
| "eval_runtime": 1697.953, | |
| "eval_samples_per_second": 20.962, | |
| "eval_steps_per_second": 0.328, | |
| "step": 6000 | |
| } | |
| ], | |
| "max_steps": 7923, | |
| "num_train_epochs": 3, | |
| "total_flos": 9.200694994360966e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |