| { | |
| "best_metric": 1.251373291015625, | |
| "best_model_checkpoint": "./saved_checkpoints/ethical/mistral/checkpoint-50", | |
| "epoch": 3.000749962501875, | |
| "eval_steps": 50, | |
| "global_step": 2500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 1.251373291015625, | |
| "eval_runtime": 4559.2323, | |
| "eval_samples_per_second": 2.193, | |
| "eval_steps_per_second": 0.366, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.842414742769675e-06, | |
| "loss": 1.2609, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 1.2726035118103027, | |
| "eval_runtime": 4449.7033, | |
| "eval_samples_per_second": 2.247, | |
| "eval_steps_per_second": 0.375, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_loss": 1.2832562923431396, | |
| "eval_runtime": 4453.6157, | |
| "eval_samples_per_second": 2.245, | |
| "eval_steps_per_second": 0.374, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2778, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 1.2775572538375854, | |
| "eval_runtime": 4472.9332, | |
| "eval_samples_per_second": 2.236, | |
| "eval_steps_per_second": 0.373, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 1.271767258644104, | |
| "eval_runtime": 4524.5689, | |
| "eval_samples_per_second": 2.21, | |
| "eval_steps_per_second": 0.368, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2742, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 1.2736577987670898, | |
| "eval_runtime": 4552.8861, | |
| "eval_samples_per_second": 2.196, | |
| "eval_steps_per_second": 0.366, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 1.2683041095733643, | |
| "eval_runtime": 4481.4789, | |
| "eval_samples_per_second": 2.231, | |
| "eval_steps_per_second": 0.372, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2707, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 1.2667808532714844, | |
| "eval_runtime": 4512.9429, | |
| "eval_samples_per_second": 2.216, | |
| "eval_steps_per_second": 0.369, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_loss": 1.2664167881011963, | |
| "eval_runtime": 4503.4031, | |
| "eval_samples_per_second": 2.221, | |
| "eval_steps_per_second": 0.37, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2633, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 1.2625447511672974, | |
| "eval_runtime": 4523.2077, | |
| "eval_samples_per_second": 2.211, | |
| "eval_steps_per_second": 0.369, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 1.2571557760238647, | |
| "eval_runtime": 4492.8054, | |
| "eval_samples_per_second": 2.226, | |
| "eval_steps_per_second": 0.371, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2605, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_loss": 1.2611154317855835, | |
| "eval_runtime": 4465.1303, | |
| "eval_samples_per_second": 2.24, | |
| "eval_steps_per_second": 0.373, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_loss": 1.261015772819519, | |
| "eval_runtime": 4573.2523, | |
| "eval_samples_per_second": 2.187, | |
| "eval_steps_per_second": 0.365, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2533, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 1.2556573152542114, | |
| "eval_runtime": 4450.2442, | |
| "eval_samples_per_second": 2.247, | |
| "eval_steps_per_second": 0.375, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 1.2533992528915405, | |
| "eval_runtime": 4528.9739, | |
| "eval_samples_per_second": 2.208, | |
| "eval_steps_per_second": 0.368, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2519, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 1.2552576065063477, | |
| "eval_runtime": 4553.5444, | |
| "eval_samples_per_second": 2.196, | |
| "eval_steps_per_second": 0.366, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 1.3253833055496216, | |
| "eval_runtime": 4454.7877, | |
| "eval_samples_per_second": 2.245, | |
| "eval_steps_per_second": 0.374, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7228, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "eval_loss": 1.3570994138717651, | |
| "eval_runtime": 4515.2615, | |
| "eval_samples_per_second": 2.215, | |
| "eval_steps_per_second": 0.369, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_loss": 1.3670397996902466, | |
| "eval_runtime": 4433.2773, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 0.376, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7433, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 1.3700823783874512, | |
| "eval_runtime": 4573.0757, | |
| "eval_samples_per_second": 2.187, | |
| "eval_steps_per_second": 0.365, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_loss": 1.36603844165802, | |
| "eval_runtime": 4525.6502, | |
| "eval_samples_per_second": 2.21, | |
| "eval_steps_per_second": 0.368, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 1e-05, | |
| "loss": 0.758, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_loss": 1.3708640336990356, | |
| "eval_runtime": 4539.0904, | |
| "eval_samples_per_second": 2.203, | |
| "eval_steps_per_second": 0.367, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 1.3683034181594849, | |
| "eval_runtime": 4412.3439, | |
| "eval_samples_per_second": 2.266, | |
| "eval_steps_per_second": 0.378, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7668, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "eval_loss": 1.3628712892532349, | |
| "eval_runtime": 4468.3564, | |
| "eval_samples_per_second": 2.238, | |
| "eval_steps_per_second": 0.373, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_loss": 1.371540904045105, | |
| "eval_runtime": 4466.1065, | |
| "eval_samples_per_second": 2.239, | |
| "eval_steps_per_second": 0.373, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7754, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_loss": 1.37712562084198, | |
| "eval_runtime": 4422.9951, | |
| "eval_samples_per_second": 2.261, | |
| "eval_steps_per_second": 0.377, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "eval_loss": 1.3581366539001465, | |
| "eval_runtime": 4440.3874, | |
| "eval_samples_per_second": 2.252, | |
| "eval_steps_per_second": 0.375, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7827, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 1.3591225147247314, | |
| "eval_runtime": 4566.96, | |
| "eval_samples_per_second": 2.19, | |
| "eval_steps_per_second": 0.365, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_loss": 1.3655798435211182, | |
| "eval_runtime": 4518.9623, | |
| "eval_samples_per_second": 2.213, | |
| "eval_steps_per_second": 0.369, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7928, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_loss": 1.3691887855529785, | |
| "eval_runtime": 4550.0865, | |
| "eval_samples_per_second": 2.198, | |
| "eval_steps_per_second": 0.366, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_loss": 1.3695429563522339, | |
| "eval_runtime": 4559.5122, | |
| "eval_samples_per_second": 2.193, | |
| "eval_steps_per_second": 0.366, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7998, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": 1.3639189004898071, | |
| "eval_runtime": 4389.9035, | |
| "eval_samples_per_second": 2.278, | |
| "eval_steps_per_second": 0.38, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_loss": 1.3616678714752197, | |
| "eval_runtime": 4562.3432, | |
| "eval_samples_per_second": 2.192, | |
| "eval_steps_per_second": 0.365, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5729, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 1.5398352146148682, | |
| "eval_runtime": 4441.3682, | |
| "eval_samples_per_second": 2.252, | |
| "eval_steps_per_second": 0.375, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_loss": 1.562680721282959, | |
| "eval_runtime": 4492.8359, | |
| "eval_samples_per_second": 2.226, | |
| "eval_steps_per_second": 0.371, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4759, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "eval_loss": 1.5819048881530762, | |
| "eval_runtime": 4460.5449, | |
| "eval_samples_per_second": 2.242, | |
| "eval_steps_per_second": 0.374, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_loss": 1.5582789182662964, | |
| "eval_runtime": 4553.8843, | |
| "eval_samples_per_second": 2.196, | |
| "eval_steps_per_second": 0.366, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4857, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "eval_loss": 1.54148530960083, | |
| "eval_runtime": 4406.6155, | |
| "eval_samples_per_second": 2.269, | |
| "eval_steps_per_second": 0.378, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "eval_loss": 1.564630150794983, | |
| "eval_runtime": 4557.1018, | |
| "eval_samples_per_second": 2.194, | |
| "eval_steps_per_second": 0.366, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4921, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 1.5717904567718506, | |
| "eval_runtime": 4430.1088, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.376, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "eval_loss": 1.5744233131408691, | |
| "eval_runtime": 4555.2722, | |
| "eval_samples_per_second": 2.195, | |
| "eval_steps_per_second": 0.366, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5034, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "eval_loss": 1.5743005275726318, | |
| "eval_runtime": 4515.3853, | |
| "eval_samples_per_second": 2.215, | |
| "eval_steps_per_second": 0.369, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "eval_loss": 1.5678597688674927, | |
| "eval_runtime": 4504.4169, | |
| "eval_samples_per_second": 2.22, | |
| "eval_steps_per_second": 0.37, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5071, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "eval_loss": 1.5610437393188477, | |
| "eval_runtime": 4513.3469, | |
| "eval_samples_per_second": 2.216, | |
| "eval_steps_per_second": 0.369, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_loss": 1.544044852256775, | |
| "eval_runtime": 4583.4964, | |
| "eval_samples_per_second": 2.182, | |
| "eval_steps_per_second": 0.364, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5117, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_loss": 1.570798397064209, | |
| "eval_runtime": 4561.3159, | |
| "eval_samples_per_second": 2.192, | |
| "eval_steps_per_second": 0.365, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "eval_loss": 1.5632375478744507, | |
| "eval_runtime": 4405.4774, | |
| "eval_samples_per_second": 2.27, | |
| "eval_steps_per_second": 0.378, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5191, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_loss": 1.5500311851501465, | |
| "eval_runtime": 4512.5389, | |
| "eval_samples_per_second": 2.216, | |
| "eval_steps_per_second": 0.369, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "eval_loss": 1.571359395980835, | |
| "eval_runtime": 4476.4923, | |
| "eval_samples_per_second": 2.234, | |
| "eval_steps_per_second": 0.372, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1e-05, | |
| "loss": 0.52, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.581729531288147, | |
| "eval_runtime": 4430.015, | |
| "eval_samples_per_second": 2.257, | |
| "eval_steps_per_second": 0.376, | |
| "step": 2500 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1000000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1201, | |
| "save_steps": 50, | |
| "total_flos": 27154959237120.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |