| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 30.0, | |
| "global_step": 6480, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 5.858560562133789, | |
| "eval_runtime": 5.0347, | |
| "eval_samples_per_second": 2349.702, | |
| "eval_steps_per_second": 2.383, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 5.509454727172852, | |
| "eval_runtime": 4.6188, | |
| "eval_samples_per_second": 2561.29, | |
| "eval_steps_per_second": 2.598, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.0001, | |
| "loss": 6.688, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 5.397603511810303, | |
| "eval_runtime": 4.6394, | |
| "eval_samples_per_second": 2549.872, | |
| "eval_steps_per_second": 2.587, | |
| "step": 648 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 5.356218338012695, | |
| "eval_runtime": 4.6644, | |
| "eval_samples_per_second": 2536.231, | |
| "eval_steps_per_second": 2.573, | |
| "step": 864 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 0.0002, | |
| "loss": 5.3629, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 5.291167736053467, | |
| "eval_runtime": 4.6079, | |
| "eval_samples_per_second": 2567.32, | |
| "eval_steps_per_second": 2.604, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 5.238525867462158, | |
| "eval_runtime": 4.628, | |
| "eval_samples_per_second": 2556.182, | |
| "eval_steps_per_second": 2.593, | |
| "step": 1296 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 0.00018175182481751826, | |
| "loss": 5.22, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 5.195512771606445, | |
| "eval_runtime": 4.6676, | |
| "eval_samples_per_second": 2534.484, | |
| "eval_steps_per_second": 2.571, | |
| "step": 1512 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 5.178501605987549, | |
| "eval_runtime": 4.6613, | |
| "eval_samples_per_second": 2537.906, | |
| "eval_steps_per_second": 2.574, | |
| "step": 1728 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 5.132693767547607, | |
| "eval_runtime": 4.6567, | |
| "eval_samples_per_second": 2540.411, | |
| "eval_steps_per_second": 2.577, | |
| "step": 1944 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "learning_rate": 0.0001635036496350365, | |
| "loss": 5.1248, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 5.1242523193359375, | |
| "eval_runtime": 4.6307, | |
| "eval_samples_per_second": 2554.696, | |
| "eval_steps_per_second": 2.591, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 5.0888824462890625, | |
| "eval_runtime": 4.6284, | |
| "eval_samples_per_second": 2555.934, | |
| "eval_steps_per_second": 2.593, | |
| "step": 2376 | |
| }, | |
| { | |
| "epoch": 11.57, | |
| "learning_rate": 0.00014525547445255475, | |
| "loss": 5.0591, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 5.073211669921875, | |
| "eval_runtime": 4.7033, | |
| "eval_samples_per_second": 2515.279, | |
| "eval_steps_per_second": 2.551, | |
| "step": 2592 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 5.041653633117676, | |
| "eval_runtime": 4.6143, | |
| "eval_samples_per_second": 2563.795, | |
| "eval_steps_per_second": 2.601, | |
| "step": 2808 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "learning_rate": 0.000127007299270073, | |
| "loss": 5.0094, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 5.038762092590332, | |
| "eval_runtime": 4.6173, | |
| "eval_samples_per_second": 2562.095, | |
| "eval_steps_per_second": 2.599, | |
| "step": 3024 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 4.9298810958862305, | |
| "eval_runtime": 4.5819, | |
| "eval_samples_per_second": 2581.88, | |
| "eval_steps_per_second": 2.619, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 4.299057960510254, | |
| "eval_runtime": 4.6127, | |
| "eval_samples_per_second": 2564.64, | |
| "eval_steps_per_second": 2.601, | |
| "step": 3456 | |
| }, | |
| { | |
| "epoch": 16.2, | |
| "learning_rate": 0.00010875912408759123, | |
| "loss": 4.7527, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 3.654055118560791, | |
| "eval_runtime": 4.5888, | |
| "eval_samples_per_second": 2578.023, | |
| "eval_steps_per_second": 2.615, | |
| "step": 3672 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 2.7825753688812256, | |
| "eval_runtime": 4.6393, | |
| "eval_samples_per_second": 2549.967, | |
| "eval_steps_per_second": 2.587, | |
| "step": 3888 | |
| }, | |
| { | |
| "epoch": 18.52, | |
| "learning_rate": 9.051094890510949e-05, | |
| "loss": 3.4431, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 2.2795569896698, | |
| "eval_runtime": 4.647, | |
| "eval_samples_per_second": 2545.709, | |
| "eval_steps_per_second": 2.582, | |
| "step": 4104 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 2.021310806274414, | |
| "eval_runtime": 4.6303, | |
| "eval_samples_per_second": 2554.922, | |
| "eval_steps_per_second": 2.592, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 20.83, | |
| "learning_rate": 7.226277372262774e-05, | |
| "loss": 2.2803, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 1.8808549642562866, | |
| "eval_runtime": 4.6167, | |
| "eval_samples_per_second": 2562.421, | |
| "eval_steps_per_second": 2.599, | |
| "step": 4536 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 1.7615374326705933, | |
| "eval_runtime": 4.6259, | |
| "eval_samples_per_second": 2557.316, | |
| "eval_steps_per_second": 2.594, | |
| "step": 4752 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 1.6925297975540161, | |
| "eval_runtime": 4.6567, | |
| "eval_samples_per_second": 2540.444, | |
| "eval_steps_per_second": 2.577, | |
| "step": 4968 | |
| }, | |
| { | |
| "epoch": 23.15, | |
| "learning_rate": 5.401459854014599e-05, | |
| "loss": 1.8601, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 1.6204941272735596, | |
| "eval_runtime": 4.7455, | |
| "eval_samples_per_second": 2492.914, | |
| "eval_steps_per_second": 2.529, | |
| "step": 5184 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 1.5750768184661865, | |
| "eval_runtime": 4.6733, | |
| "eval_samples_per_second": 2531.399, | |
| "eval_steps_per_second": 2.568, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 25.46, | |
| "learning_rate": 3.5766423357664236e-05, | |
| "loss": 1.6697, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 1.5390561819076538, | |
| "eval_runtime": 4.6271, | |
| "eval_samples_per_second": 2556.66, | |
| "eval_steps_per_second": 2.593, | |
| "step": 5616 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 1.520015835762024, | |
| "eval_runtime": 4.6894, | |
| "eval_samples_per_second": 2522.687, | |
| "eval_steps_per_second": 2.559, | |
| "step": 5832 | |
| }, | |
| { | |
| "epoch": 27.78, | |
| "learning_rate": 1.7518248175182482e-05, | |
| "loss": 1.5655, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 1.4865714311599731, | |
| "eval_runtime": 4.6379, | |
| "eval_samples_per_second": 2550.726, | |
| "eval_steps_per_second": 2.587, | |
| "step": 6048 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 1.4655797481536865, | |
| "eval_runtime": 4.6124, | |
| "eval_samples_per_second": 2564.811, | |
| "eval_steps_per_second": 2.602, | |
| "step": 6264 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 1.4627275466918945, | |
| "eval_runtime": 4.9986, | |
| "eval_samples_per_second": 2366.681, | |
| "eval_steps_per_second": 2.401, | |
| "step": 6480 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 6480, | |
| "total_flos": 5.459072646905856e+16, | |
| "train_loss": 3.818386595926167, | |
| "train_runtime": 5045.7987, | |
| "train_samples_per_second": 1310.118, | |
| "train_steps_per_second": 1.284 | |
| } | |
| ], | |
| "max_steps": 6480, | |
| "num_train_epochs": 30, | |
| "total_flos": 5.459072646905856e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |