| { | |
| "best_metric": 0.28675174713134766, | |
| "best_model_checkpoint": "ckpts/discrim/contrastive_models/gen.llama-7b/flan_t5_large_coin_flip_100K_simcse_fixed_align/checkpoint-1500", | |
| "epoch": 1.337196344996657, | |
| "global_step": 1500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.457652303120357e-06, | |
| "loss": 0.9997, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 8.915304606240713e-06, | |
| "loss": 0.9994, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.337295690936107e-05, | |
| "loss": 0.9958, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.7830609212481426e-05, | |
| "loss": 0.9517, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.2288261515601784e-05, | |
| "loss": 0.7892, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 2.674591381872214e-05, | |
| "loss": 0.7113, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 3.12035661218425e-05, | |
| "loss": 0.6393, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 3.566121842496285e-05, | |
| "loss": 0.6115, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.011887072808321e-05, | |
| "loss": 0.5684, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.457652303120357e-05, | |
| "loss": 0.4807, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_loss": 0.39063507318496704, | |
| "eval_runtime": 46.0326, | |
| "eval_samples_per_second": 132.819, | |
| "eval_steps_per_second": 16.619, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.903417533432392e-05, | |
| "loss": 0.4024, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 5.349182763744428e-05, | |
| "loss": 0.3846, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 5.794947994056464e-05, | |
| "loss": 0.336, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 5.9999027964273596e-05, | |
| "loss": 0.3333, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 5.9992094680623114e-05, | |
| "loss": 0.3155, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 5.9978496329511386e-05, | |
| "loss": 0.3047, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 5.995823593286565e-05, | |
| "loss": 0.257, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 5.993131799310269e-05, | |
| "loss": 0.2651, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 5.989774849212826e-05, | |
| "loss": 0.2447, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 5.9857534890007815e-05, | |
| "loss": 0.2189, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_loss": 0.30103570222854614, | |
| "eval_runtime": 46.0873, | |
| "eval_samples_per_second": 132.661, | |
| "eval_steps_per_second": 16.599, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 5.981068612330859e-05, | |
| "loss": 0.2223, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 5.975721260311371e-05, | |
| "loss": 0.21, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 5.969712621270851e-05, | |
| "loss": 0.1952, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 5.963044030493981e-05, | |
| "loss": 0.2101, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 5.955716969924848e-05, | |
| "loss": 0.1969, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 5.947733067837621e-05, | |
| "loss": 0.1706, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 5.939094098474696e-05, | |
| "loss": 0.1491, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 5.929801981652421e-05, | |
| "loss": 0.1209, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 5.9198587823344496e-05, | |
| "loss": 0.1119, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 5.909266710172858e-05, | |
| "loss": 0.1188, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "eval_loss": 0.28675174713134766, | |
| "eval_runtime": 46.0465, | |
| "eval_samples_per_second": 132.779, | |
| "eval_steps_per_second": 16.614, | |
| "step": 1500 | |
| } | |
| ], | |
| "max_steps": 11210, | |
| "num_train_epochs": 10, | |
| "total_flos": 0.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |