| { | |
| "best_metric": 1.110647439956665, | |
| "best_model_checkpoint": "outs/instuct_chat_50k/checkpoint-2000", | |
| "epoch": 0.23353573096683794, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1.5568240788790864e-07, | |
| "loss": 1.1388, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.5412558380902958e-05, | |
| "loss": 1.5551, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_loss": 1.2936956882476807, | |
| "eval_runtime": 31.4406, | |
| "eval_samples_per_second": 6.361, | |
| "eval_steps_per_second": 0.795, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 3.098079916969382e-05, | |
| "loss": 1.4423, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "eval_loss": 1.227220892906189, | |
| "eval_runtime": 31.4526, | |
| "eval_samples_per_second": 6.359, | |
| "eval_steps_per_second": 0.795, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 4.6393357550596784e-05, | |
| "loss": 1.3335, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "eval_loss": 1.2049953937530518, | |
| "eval_runtime": 31.9522, | |
| "eval_samples_per_second": 6.259, | |
| "eval_steps_per_second": 0.782, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 6.165023352361183e-05, | |
| "loss": 1.4225, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_loss": 1.1863670349121094, | |
| "eval_runtime": 31.5148, | |
| "eval_samples_per_second": 6.346, | |
| "eval_steps_per_second": 0.793, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 7.721847431240269e-05, | |
| "loss": 1.3618, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_loss": 1.1773875951766968, | |
| "eval_runtime": 31.4521, | |
| "eval_samples_per_second": 6.359, | |
| "eval_steps_per_second": 0.795, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.278671510119357e-05, | |
| "loss": 1.3262, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 1.168226718902588, | |
| "eval_runtime": 31.5121, | |
| "eval_samples_per_second": 6.347, | |
| "eval_steps_per_second": 0.793, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 0.00010835495588998442, | |
| "loss": 1.2133, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 1.1481703519821167, | |
| "eval_runtime": 31.504, | |
| "eval_samples_per_second": 6.348, | |
| "eval_steps_per_second": 0.794, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00012392319667877528, | |
| "loss": 1.2034, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 1.13801908493042, | |
| "eval_runtime": 31.5183, | |
| "eval_samples_per_second": 6.346, | |
| "eval_steps_per_second": 0.793, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.00013949143746756616, | |
| "loss": 1.2159, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 1.1303695440292358, | |
| "eval_runtime": 31.4531, | |
| "eval_samples_per_second": 6.359, | |
| "eval_steps_per_second": 0.795, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00015505967825635704, | |
| "loss": 1.1612, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 1.1280632019042969, | |
| "eval_runtime": 31.434, | |
| "eval_samples_per_second": 6.363, | |
| "eval_steps_per_second": 0.795, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00017062791904514786, | |
| "loss": 1.1549, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_loss": 1.1211422681808472, | |
| "eval_runtime": 31.5065, | |
| "eval_samples_per_second": 6.348, | |
| "eval_steps_per_second": 0.793, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00018619615983393875, | |
| "loss": 1.2038, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_loss": 1.126134991645813, | |
| "eval_runtime": 31.4919, | |
| "eval_samples_per_second": 6.351, | |
| "eval_steps_per_second": 0.794, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 0.00020176440062272963, | |
| "loss": 1.1784, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 1.1223646402359009, | |
| "eval_runtime": 31.4108, | |
| "eval_samples_per_second": 6.367, | |
| "eval_steps_per_second": 0.796, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00021733264141152048, | |
| "loss": 1.1834, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_loss": 1.113663911819458, | |
| "eval_runtime": 31.463, | |
| "eval_samples_per_second": 6.357, | |
| "eval_steps_per_second": 0.795, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00023290088220031133, | |
| "loss": 1.1974, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 1.1265443563461304, | |
| "eval_runtime": 31.4893, | |
| "eval_samples_per_second": 6.351, | |
| "eval_steps_per_second": 0.794, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0002484691229891022, | |
| "loss": 1.2174, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 1.1237465143203735, | |
| "eval_runtime": 31.5226, | |
| "eval_samples_per_second": 6.345, | |
| "eval_steps_per_second": 0.793, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00026403736377789307, | |
| "loss": 1.2019, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 1.1198475360870361, | |
| "eval_runtime": 31.4658, | |
| "eval_samples_per_second": 6.356, | |
| "eval_steps_per_second": 0.795, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.0002796056045666839, | |
| "loss": 1.2114, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_loss": 1.1201978921890259, | |
| "eval_runtime": 31.5138, | |
| "eval_samples_per_second": 6.346, | |
| "eval_steps_per_second": 0.793, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00029517384535547483, | |
| "loss": 1.1848, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 1.1198968887329102, | |
| "eval_runtime": 31.1932, | |
| "eval_samples_per_second": 6.412, | |
| "eval_steps_per_second": 0.801, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.0002994345961596241, | |
| "loss": 1.129, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 1.110647439956665, | |
| "eval_runtime": 31.1909, | |
| "eval_samples_per_second": 6.412, | |
| "eval_steps_per_second": 0.802, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.00029861517030400697, | |
| "loss": 1.1864, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 1.1134731769561768, | |
| "eval_runtime": 31.169, | |
| "eval_samples_per_second": 6.417, | |
| "eval_steps_per_second": 0.802, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0002977957444483898, | |
| "loss": 1.1725, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_loss": 1.1226890087127686, | |
| "eval_runtime": 31.1975, | |
| "eval_samples_per_second": 6.411, | |
| "eval_steps_per_second": 0.801, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0002969763185927726, | |
| "loss": 1.1363, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_loss": 1.1255100965499878, | |
| "eval_runtime": 31.237, | |
| "eval_samples_per_second": 6.403, | |
| "eval_steps_per_second": 0.8, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00029615689273715547, | |
| "loss": 1.204, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 1.1148998737335205, | |
| "eval_runtime": 31.2369, | |
| "eval_samples_per_second": 6.403, | |
| "eval_steps_per_second": 0.8, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0002953374668815383, | |
| "loss": 1.1465, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 1.1103030443191528, | |
| "eval_runtime": 31.2141, | |
| "eval_samples_per_second": 6.407, | |
| "eval_steps_per_second": 0.801, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0002945180410259212, | |
| "loss": 1.2058, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 1.1112116575241089, | |
| "eval_runtime": 31.6225, | |
| "eval_samples_per_second": 6.325, | |
| "eval_steps_per_second": 0.791, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00029369861517030397, | |
| "loss": 1.1662, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 1.1122115850448608, | |
| "eval_runtime": 31.1829, | |
| "eval_samples_per_second": 6.414, | |
| "eval_steps_per_second": 0.802, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002928791893146868, | |
| "loss": 1.1372, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 1.1113629341125488, | |
| "eval_runtime": 31.6588, | |
| "eval_samples_per_second": 6.317, | |
| "eval_steps_per_second": 0.79, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0002920597634590696, | |
| "loss": 1.1693, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_loss": 1.1073808670043945, | |
| "eval_runtime": 31.5886, | |
| "eval_samples_per_second": 6.331, | |
| "eval_steps_per_second": 0.791, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0002912403376034525, | |
| "loss": 1.1776, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_loss": 1.1112266778945923, | |
| "eval_runtime": 31.3957, | |
| "eval_samples_per_second": 6.37, | |
| "eval_steps_per_second": 0.796, | |
| "step": 3000 | |
| } | |
| ], | |
| "max_steps": 38538, | |
| "num_train_epochs": 3, | |
| "total_flos": 4.679774978918646e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |