| { | |
| "best_metric": 1.2135709524154663, | |
| "best_model_checkpoint": "./outputs/checkpoint-4000", | |
| "epoch": 2.9143897996357016, | |
| "eval_steps": 100, | |
| "global_step": 4000, | |
| "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": 2.1514, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 1.9972127676010132, | |
| "eval_runtime": 132.9222, | |
| "eval_samples_per_second": 47.201, | |
| "eval_steps_per_second": 5.906, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 1.9494, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 1.9086177349090576, | |
| "eval_runtime": 132.9394, | |
| "eval_samples_per_second": 47.194, | |
| "eval_steps_per_second": 5.905, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8786, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 1.8495125770568848, | |
| "eval_runtime": 132.8956, | |
| "eval_samples_per_second": 47.21, | |
| "eval_steps_per_second": 5.907, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 1.8283, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 1.8054330348968506, | |
| "eval_runtime": 132.877, | |
| "eval_samples_per_second": 47.217, | |
| "eval_steps_per_second": 5.908, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7777, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 1.7643086910247803, | |
| "eval_runtime": 132.9187, | |
| "eval_samples_per_second": 47.202, | |
| "eval_steps_per_second": 5.906, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 1.7496, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.7355191707611084, | |
| "eval_runtime": 132.8029, | |
| "eval_samples_per_second": 47.243, | |
| "eval_steps_per_second": 5.911, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 1.712, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 1.703617811203003, | |
| "eval_runtime": 132.9047, | |
| "eval_samples_per_second": 47.207, | |
| "eval_steps_per_second": 5.906, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 1.691, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 1.675934076309204, | |
| "eval_runtime": 132.9001, | |
| "eval_samples_per_second": 47.208, | |
| "eval_steps_per_second": 5.907, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6552, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 1.653730869293213, | |
| "eval_runtime": 132.8143, | |
| "eval_samples_per_second": 47.239, | |
| "eval_steps_per_second": 5.911, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6412, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 1.629328966140747, | |
| "eval_runtime": 132.9175, | |
| "eval_samples_per_second": 47.202, | |
| "eval_steps_per_second": 5.906, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 1.6211, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.6079449653625488, | |
| "eval_runtime": 132.8611, | |
| "eval_samples_per_second": 47.222, | |
| "eval_steps_per_second": 5.908, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5861, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 1.5884170532226562, | |
| "eval_runtime": 132.8296, | |
| "eval_samples_per_second": 47.233, | |
| "eval_steps_per_second": 5.91, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5925, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 1.5686501264572144, | |
| "eval_runtime": 132.8517, | |
| "eval_samples_per_second": 47.226, | |
| "eval_steps_per_second": 5.909, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5467, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 1.5463733673095703, | |
| "eval_runtime": 132.8859, | |
| "eval_samples_per_second": 47.213, | |
| "eval_steps_per_second": 5.907, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5182, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 1.5295236110687256, | |
| "eval_runtime": 132.8424, | |
| "eval_samples_per_second": 47.229, | |
| "eval_steps_per_second": 5.909, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002, | |
| "loss": 1.5007, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 1.5132012367248535, | |
| "eval_runtime": 132.7525, | |
| "eval_samples_per_second": 47.261, | |
| "eval_steps_per_second": 5.913, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4987, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 1.496045708656311, | |
| "eval_runtime": 132.8633, | |
| "eval_samples_per_second": 47.221, | |
| "eval_steps_per_second": 5.908, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0002, | |
| "loss": 1.471, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 1.4813148975372314, | |
| "eval_runtime": 132.8337, | |
| "eval_samples_per_second": 47.232, | |
| "eval_steps_per_second": 5.91, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4629, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 1.4688937664031982, | |
| "eval_runtime": 132.906, | |
| "eval_samples_per_second": 47.206, | |
| "eval_steps_per_second": 5.906, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4448, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 1.4537802934646606, | |
| "eval_runtime": 132.7872, | |
| "eval_samples_per_second": 47.249, | |
| "eval_steps_per_second": 5.912, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4308, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 1.437719464302063, | |
| "eval_runtime": 132.854, | |
| "eval_samples_per_second": 47.225, | |
| "eval_steps_per_second": 5.909, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4061, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 1.423782467842102, | |
| "eval_runtime": 132.744, | |
| "eval_samples_per_second": 47.264, | |
| "eval_steps_per_second": 5.914, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4041, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 1.4104719161987305, | |
| "eval_runtime": 132.8625, | |
| "eval_samples_per_second": 47.222, | |
| "eval_steps_per_second": 5.908, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0002, | |
| "loss": 1.404, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 1.3974003791809082, | |
| "eval_runtime": 132.8724, | |
| "eval_samples_per_second": 47.218, | |
| "eval_steps_per_second": 5.908, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3756, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 1.3845711946487427, | |
| "eval_runtime": 132.8814, | |
| "eval_samples_per_second": 47.215, | |
| "eval_steps_per_second": 5.908, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3654, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 1.3717011213302612, | |
| "eval_runtime": 132.776, | |
| "eval_samples_per_second": 47.253, | |
| "eval_steps_per_second": 5.912, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3535, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 1.3580347299575806, | |
| "eval_runtime": 132.7648, | |
| "eval_samples_per_second": 47.256, | |
| "eval_steps_per_second": 5.913, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3256, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 1.3460556268692017, | |
| "eval_runtime": 132.8473, | |
| "eval_samples_per_second": 47.227, | |
| "eval_steps_per_second": 5.909, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002, | |
| "loss": 1.299, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 1.333855152130127, | |
| "eval_runtime": 132.9091, | |
| "eval_samples_per_second": 47.205, | |
| "eval_steps_per_second": 5.906, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3002, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_loss": 1.3230880498886108, | |
| "eval_runtime": 132.8618, | |
| "eval_samples_per_second": 47.222, | |
| "eval_steps_per_second": 5.908, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2797, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_loss": 1.3130332231521606, | |
| "eval_runtime": 132.7387, | |
| "eval_samples_per_second": 47.266, | |
| "eval_steps_per_second": 5.914, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2834, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 1.3008962869644165, | |
| "eval_runtime": 132.866, | |
| "eval_samples_per_second": 47.221, | |
| "eval_steps_per_second": 5.908, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0002, | |
| "loss": 1.266, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 1.28789484500885, | |
| "eval_runtime": 132.8174, | |
| "eval_samples_per_second": 47.238, | |
| "eval_steps_per_second": 5.91, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2491, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "eval_loss": 1.2794678211212158, | |
| "eval_runtime": 132.939, | |
| "eval_samples_per_second": 47.195, | |
| "eval_steps_per_second": 5.905, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2486, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 1.2680211067199707, | |
| "eval_runtime": 132.8824, | |
| "eval_samples_per_second": 47.215, | |
| "eval_steps_per_second": 5.907, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2372, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 1.2562363147735596, | |
| "eval_runtime": 132.9828, | |
| "eval_samples_per_second": 47.179, | |
| "eval_steps_per_second": 5.903, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2199, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_loss": 1.2433452606201172, | |
| "eval_runtime": 132.8033, | |
| "eval_samples_per_second": 47.243, | |
| "eval_steps_per_second": 5.911, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1973, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_loss": 1.233286738395691, | |
| "eval_runtime": 132.8372, | |
| "eval_samples_per_second": 47.231, | |
| "eval_steps_per_second": 5.909, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2073, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_loss": 1.2219091653823853, | |
| "eval_runtime": 132.7605, | |
| "eval_samples_per_second": 47.258, | |
| "eval_steps_per_second": 5.913, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 0.0002, | |
| "loss": 1.1906, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_loss": 1.2135709524154663, | |
| "eval_runtime": 132.8138, | |
| "eval_samples_per_second": 47.239, | |
| "eval_steps_per_second": 5.911, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 2.3840611478920397e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |