| { | |
| "best_metric": 0.9393565058708191, | |
| "best_model_checkpoint": "./checkpoint-220", | |
| "epoch": 5.930232558139535, | |
| "global_step": 220, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 5e-05, | |
| "loss": 6.7937, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 6.501975059509277, | |
| "eval_runtime": 33.9782, | |
| "eval_samples_per_second": 59.685, | |
| "eval_steps_per_second": 1.884, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0001, | |
| "loss": 5.985, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": 5.326801776885986, | |
| "eval_runtime": 18.802, | |
| "eval_samples_per_second": 107.861, | |
| "eval_steps_per_second": 3.404, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.00015, | |
| "loss": 4.6798, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 3.829317331314087, | |
| "eval_runtime": 18.7742, | |
| "eval_samples_per_second": 108.021, | |
| "eval_steps_per_second": 3.409, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.0002, | |
| "loss": 3.0462, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "eval_loss": 1.9157757759094238, | |
| "eval_runtime": 18.7907, | |
| "eval_samples_per_second": 107.926, | |
| "eval_steps_per_second": 3.406, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.00025, | |
| "loss": 1.5197, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "eval_loss": 1.2286747694015503, | |
| "eval_runtime": 18.812, | |
| "eval_samples_per_second": 107.804, | |
| "eval_steps_per_second": 3.402, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.0003, | |
| "loss": 1.1349, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "eval_loss": 1.1046648025512695, | |
| "eval_runtime": 18.8413, | |
| "eval_samples_per_second": 107.636, | |
| "eval_steps_per_second": 3.397, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.00035, | |
| "loss": 1.0718, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "eval_loss": 1.0654218196868896, | |
| "eval_runtime": 18.8231, | |
| "eval_samples_per_second": 107.74, | |
| "eval_steps_per_second": 3.4, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 0.0004, | |
| "loss": 1.0987, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "eval_loss": 1.0404733419418335, | |
| "eval_runtime": 18.8082, | |
| "eval_samples_per_second": 107.825, | |
| "eval_steps_per_second": 3.403, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 1.0133, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "eval_loss": 1.0208371877670288, | |
| "eval_runtime": 18.8192, | |
| "eval_samples_per_second": 107.762, | |
| "eval_steps_per_second": 3.401, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.0005, | |
| "loss": 0.9869, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "eval_loss": 1.0047191381454468, | |
| "eval_runtime": 18.852, | |
| "eval_samples_per_second": 107.575, | |
| "eval_steps_per_second": 3.395, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 0.00045901639344262296, | |
| "loss": 0.9809, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_loss": 0.9912181496620178, | |
| "eval_runtime": 18.8516, | |
| "eval_samples_per_second": 107.577, | |
| "eval_steps_per_second": 3.395, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 0.0004180327868852459, | |
| "loss": 1.0275, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "eval_loss": 0.9803428649902344, | |
| "eval_runtime": 18.8293, | |
| "eval_samples_per_second": 107.705, | |
| "eval_steps_per_second": 3.399, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 0.0003770491803278688, | |
| "loss": 0.9608, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "eval_loss": 0.9717100262641907, | |
| "eval_runtime": 18.8516, | |
| "eval_samples_per_second": 107.577, | |
| "eval_steps_per_second": 3.395, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 0.0003360655737704918, | |
| "loss": 0.9431, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "eval_loss": 0.9643934965133667, | |
| "eval_runtime": 18.8147, | |
| "eval_samples_per_second": 107.788, | |
| "eval_steps_per_second": 3.402, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 0.00029508196721311476, | |
| "loss": 0.991, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "eval_loss": 0.9581753015518188, | |
| "eval_runtime": 18.8481, | |
| "eval_samples_per_second": 107.597, | |
| "eval_steps_per_second": 3.396, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 0.0002540983606557377, | |
| "loss": 0.9387, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "eval_loss": 0.9531411528587341, | |
| "eval_runtime": 18.8022, | |
| "eval_samples_per_second": 107.86, | |
| "eval_steps_per_second": 3.404, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 0.00021311475409836064, | |
| "loss": 0.9203, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "eval_loss": 0.9489945769309998, | |
| "eval_runtime": 18.7924, | |
| "eval_samples_per_second": 107.916, | |
| "eval_steps_per_second": 3.406, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "learning_rate": 0.00017213114754098362, | |
| "loss": 0.9235, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "eval_loss": 0.9456363320350647, | |
| "eval_runtime": 18.8084, | |
| "eval_samples_per_second": 107.824, | |
| "eval_steps_per_second": 3.403, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "learning_rate": 0.00013114754098360657, | |
| "loss": 0.9746, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "eval_loss": 0.9429621696472168, | |
| "eval_runtime": 18.8357, | |
| "eval_samples_per_second": 107.668, | |
| "eval_steps_per_second": 3.398, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "learning_rate": 9.016393442622952e-05, | |
| "loss": 0.9176, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "eval_loss": 0.9410804510116577, | |
| "eval_runtime": 18.8338, | |
| "eval_samples_per_second": 107.679, | |
| "eval_steps_per_second": 3.398, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 4.9180327868852456e-05, | |
| "loss": 0.9175, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "eval_loss": 0.9398788213729858, | |
| "eval_runtime": 18.8076, | |
| "eval_samples_per_second": 107.829, | |
| "eval_steps_per_second": 3.403, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 8.19672131147541e-06, | |
| "loss": 0.91, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "eval_loss": 0.9393565058708191, | |
| "eval_runtime": 18.7996, | |
| "eval_samples_per_second": 107.874, | |
| "eval_steps_per_second": 3.404, | |
| "step": 220 | |
| } | |
| ], | |
| "max_steps": 222, | |
| "num_train_epochs": 6, | |
| "total_flos": 8.954608582656e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |