| { | |
| "best_metric": 6729.03466796875, | |
| "best_model_checkpoint": "./coco_outputs/checkpoint-640", | |
| "epoch": 5.0, | |
| "global_step": 640, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.96875e-07, | |
| "loss": 6346.4363, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.9375e-07, | |
| "loss": 6401.3863, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.90625e-07, | |
| "loss": 6297.2812, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.875e-07, | |
| "loss": 6283.0867, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 1.8437499999999998e-07, | |
| "loss": 6141.998, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.8124999999999999e-07, | |
| "loss": 6092.3895, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1.78125e-07, | |
| "loss": 6228.5742, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.75e-07, | |
| "loss": 6152.1148, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.71875e-07, | |
| "loss": 6431.5375, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 1.6875e-07, | |
| "loss": 6201.1875, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.65625e-07, | |
| "loss": 6414.7582, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 1.6249999999999998e-07, | |
| "loss": 6463.6664, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 6729.2021484375, | |
| "eval_runtime": 17.4768, | |
| "eval_samples_per_second": 14.648, | |
| "eval_steps_per_second": 1.831, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.5937499999999998e-07, | |
| "loss": 6203.2496, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 1.5624999999999999e-07, | |
| "loss": 6195.1863, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.53125e-07, | |
| "loss": 6365.1023, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1.5e-07, | |
| "loss": 6115.4832, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.46875e-07, | |
| "loss": 6507.116, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.4375e-07, | |
| "loss": 6235.0008, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 1.40625e-07, | |
| "loss": 6424.7523, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 1.375e-07, | |
| "loss": 6032.757, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 1.3437499999999998e-07, | |
| "loss": 5972.9164, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 1.3125e-07, | |
| "loss": 5884.6973, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.28125e-07, | |
| "loss": 6157.3324, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 1.25e-07, | |
| "loss": 6338.2691, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 1.21875e-07, | |
| "loss": 6242.4801, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 6729.1201171875, | |
| "eval_runtime": 17.3704, | |
| "eval_samples_per_second": 14.738, | |
| "eval_steps_per_second": 1.842, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 1.1875e-07, | |
| "loss": 6287.9883, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 1.1562499999999999e-07, | |
| "loss": 6261.4707, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.125e-07, | |
| "loss": 6337.7199, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1.09375e-07, | |
| "loss": 6193.377, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 1.0624999999999999e-07, | |
| "loss": 5926.5789, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 1.0312499999999999e-07, | |
| "loss": 6122.5051, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 1e-07, | |
| "loss": 6157.1656, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 9.6875e-08, | |
| "loss": 6162.8938, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 9.375e-08, | |
| "loss": 6154.5023, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 9.062499999999999e-08, | |
| "loss": 6107.1387, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 8.75e-08, | |
| "loss": 6092.2199, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 8.4375e-08, | |
| "loss": 6291.2375, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 8.124999999999999e-08, | |
| "loss": 6344.1949, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 6729.05859375, | |
| "eval_runtime": 17.488, | |
| "eval_samples_per_second": 14.639, | |
| "eval_steps_per_second": 1.83, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 7.812499999999999e-08, | |
| "loss": 6475.0023, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 7.5e-08, | |
| "loss": 6335.7672, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 7.1875e-08, | |
| "loss": 6222.3957, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 6.875e-08, | |
| "loss": 6346.1539, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 6.5625e-08, | |
| "loss": 6196.7348, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 6.25e-08, | |
| "loss": 6440.0867, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 5.9375e-08, | |
| "loss": 6186.3539, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 5.625e-08, | |
| "loss": 6364.8961, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 5.3124999999999994e-08, | |
| "loss": 6580.9484, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 5e-08, | |
| "loss": 6384.6832, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 4.6875e-08, | |
| "loss": 6195.5973, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 4.375e-08, | |
| "loss": 6409.9172, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 4.0624999999999995e-08, | |
| "loss": 6528.6664, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 6729.0419921875, | |
| "eval_runtime": 17.1328, | |
| "eval_samples_per_second": 14.942, | |
| "eval_steps_per_second": 1.868, | |
| "step": 512 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 3.75e-08, | |
| "loss": 6024.7227, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 3.4375e-08, | |
| "loss": 5942.5199, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 3.125e-08, | |
| "loss": 6240.5281, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 2.8125e-08, | |
| "loss": 6593.8445, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 2.5e-08, | |
| "loss": 6542.1891, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.45, | |
| "learning_rate": 2.1875e-08, | |
| "loss": 6388.3813, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "learning_rate": 1.875e-08, | |
| "loss": 6358.7457, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 1.5625e-08, | |
| "loss": 5996.3961, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 1.25e-08, | |
| "loss": 6177.6547, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 9.375e-09, | |
| "loss": 6032.8859, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 6.25e-09, | |
| "loss": 6523.9477, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 3.125e-09, | |
| "loss": 6293.6355, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 5746.1234, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 6729.03466796875, | |
| "eval_runtime": 17.3893, | |
| "eval_samples_per_second": 14.722, | |
| "eval_steps_per_second": 1.84, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 640, | |
| "total_flos": 2.4480067682304e+18, | |
| "train_loss": 6251.508282470703, | |
| "train_runtime": 838.0786, | |
| "train_samples_per_second": 6.109, | |
| "train_steps_per_second": 0.764 | |
| } | |
| ], | |
| "max_steps": 640, | |
| "num_train_epochs": 5, | |
| "total_flos": 2.4480067682304e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |