| { | |
| "best_metric": 2.0894112586975098, | |
| "best_model_checkpoint": "./outputs/checkpoint-3700", | |
| "epoch": 2.6958105646630237, | |
| "eval_steps": 100, | |
| "global_step": 3700, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0002, | |
| "loss": 2.8181, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.7240540981292725, | |
| "eval_runtime": 204.8489, | |
| "eval_samples_per_second": 30.627, | |
| "eval_steps_per_second": 3.832, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6957, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.6725997924804688, | |
| "eval_runtime": 204.875, | |
| "eval_samples_per_second": 30.624, | |
| "eval_steps_per_second": 3.832, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6509, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.6377696990966797, | |
| "eval_runtime": 204.8779, | |
| "eval_samples_per_second": 30.623, | |
| "eval_steps_per_second": 3.832, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6248, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 2.6048855781555176, | |
| "eval_runtime": 204.7551, | |
| "eval_samples_per_second": 30.641, | |
| "eval_steps_per_second": 3.834, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5822, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 2.5791709423065186, | |
| "eval_runtime": 204.976, | |
| "eval_samples_per_second": 30.608, | |
| "eval_steps_per_second": 3.83, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5557, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 2.5526647567749023, | |
| "eval_runtime": 208.5751, | |
| "eval_samples_per_second": 30.08, | |
| "eval_steps_per_second": 3.764, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5397, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 2.5280492305755615, | |
| "eval_runtime": 204.8306, | |
| "eval_samples_per_second": 30.63, | |
| "eval_steps_per_second": 3.832, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5165, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 2.507997512817383, | |
| "eval_runtime": 204.6726, | |
| "eval_samples_per_second": 30.654, | |
| "eval_steps_per_second": 3.835, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 2.478, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 2.485630512237549, | |
| "eval_runtime": 204.5609, | |
| "eval_samples_per_second": 30.671, | |
| "eval_steps_per_second": 3.837, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4748, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 2.4663589000701904, | |
| "eval_runtime": 204.9278, | |
| "eval_samples_per_second": 30.616, | |
| "eval_steps_per_second": 3.831, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4692, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 2.4438841342926025, | |
| "eval_runtime": 204.9158, | |
| "eval_samples_per_second": 30.617, | |
| "eval_steps_per_second": 3.831, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4221, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 2.42663836479187, | |
| "eval_runtime": 204.7233, | |
| "eval_samples_per_second": 30.646, | |
| "eval_steps_per_second": 3.834, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 2.436, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 2.4082868099212646, | |
| "eval_runtime": 204.6174, | |
| "eval_samples_per_second": 30.662, | |
| "eval_steps_per_second": 3.836, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3811, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 2.3840975761413574, | |
| "eval_runtime": 204.8146, | |
| "eval_samples_per_second": 30.633, | |
| "eval_steps_per_second": 3.833, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3425, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 2.3692121505737305, | |
| "eval_runtime": 204.8938, | |
| "eval_samples_per_second": 30.621, | |
| "eval_steps_per_second": 3.831, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3349, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 2.3504340648651123, | |
| "eval_runtime": 205.0412, | |
| "eval_samples_per_second": 30.599, | |
| "eval_steps_per_second": 3.828, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3358, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 2.3324663639068604, | |
| "eval_runtime": 205.0224, | |
| "eval_samples_per_second": 30.602, | |
| "eval_steps_per_second": 3.829, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2973, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 2.3180932998657227, | |
| "eval_runtime": 204.7898, | |
| "eval_samples_per_second": 30.636, | |
| "eval_steps_per_second": 3.833, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2809, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 2.302474021911621, | |
| "eval_runtime": 205.1402, | |
| "eval_samples_per_second": 30.584, | |
| "eval_steps_per_second": 3.827, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2762, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 2.28745436668396, | |
| "eval_runtime": 205.0203, | |
| "eval_samples_per_second": 30.602, | |
| "eval_steps_per_second": 3.829, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2577, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 2.273569107055664, | |
| "eval_runtime": 205.4686, | |
| "eval_samples_per_second": 30.535, | |
| "eval_steps_per_second": 3.821, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2406, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 2.258324384689331, | |
| "eval_runtime": 205.2879, | |
| "eval_samples_per_second": 30.562, | |
| "eval_steps_per_second": 3.824, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2328, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 2.246018409729004, | |
| "eval_runtime": 205.2551, | |
| "eval_samples_per_second": 30.567, | |
| "eval_steps_per_second": 3.825, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0002, | |
| "loss": 2.236, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 2.2329859733581543, | |
| "eval_runtime": 205.4741, | |
| "eval_samples_per_second": 30.534, | |
| "eval_steps_per_second": 3.82, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2046, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 2.2189931869506836, | |
| "eval_runtime": 205.4106, | |
| "eval_samples_per_second": 30.544, | |
| "eval_steps_per_second": 3.822, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2048, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 2.206812858581543, | |
| "eval_runtime": 204.865, | |
| "eval_samples_per_second": 30.625, | |
| "eval_steps_per_second": 3.832, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0002, | |
| "loss": 2.188, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 2.1901111602783203, | |
| "eval_runtime": 205.0397, | |
| "eval_samples_per_second": 30.599, | |
| "eval_steps_per_second": 3.829, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0002, | |
| "loss": 2.145, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 2.1863455772399902, | |
| "eval_runtime": 205.3424, | |
| "eval_samples_per_second": 30.554, | |
| "eval_steps_per_second": 3.823, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002, | |
| "loss": 2.116, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 2.171632766723633, | |
| "eval_runtime": 205.19, | |
| "eval_samples_per_second": 30.577, | |
| "eval_steps_per_second": 3.826, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1341, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_loss": 2.162501096725464, | |
| "eval_runtime": 204.9372, | |
| "eval_samples_per_second": 30.614, | |
| "eval_steps_per_second": 3.83, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1015, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 2.149629831314087, | |
| "eval_runtime": 205.2556, | |
| "eval_samples_per_second": 30.567, | |
| "eval_steps_per_second": 3.825, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0002, | |
| "loss": 2.122, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 2.139963388442993, | |
| "eval_runtime": 205.3965, | |
| "eval_samples_per_second": 30.546, | |
| "eval_steps_per_second": 3.822, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0835, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 2.128814935684204, | |
| "eval_runtime": 205.0918, | |
| "eval_samples_per_second": 30.591, | |
| "eval_steps_per_second": 3.828, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0783, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_loss": 2.1278133392333984, | |
| "eval_runtime": 205.2987, | |
| "eval_samples_per_second": 30.56, | |
| "eval_steps_per_second": 3.824, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0761, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 2.1079277992248535, | |
| "eval_runtime": 205.5563, | |
| "eval_samples_per_second": 30.522, | |
| "eval_steps_per_second": 3.819, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0002, | |
| "loss": 2.076, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 2.09975004196167, | |
| "eval_runtime": 206.3028, | |
| "eval_samples_per_second": 30.412, | |
| "eval_steps_per_second": 3.805, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0002, | |
| "loss": 2.0554, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_loss": 2.0894112586975098, | |
| "eval_runtime": 205.5534, | |
| "eval_samples_per_second": 30.522, | |
| "eval_steps_per_second": 3.819, | |
| "step": 3700 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 1.0533333313456128e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |