| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.996941896024465, | |
| "eval_steps": 1000, | |
| "global_step": 1470, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1019367991845056, | |
| "grad_norm": 0.2208942323923111, | |
| "learning_rate": 9.999696229471716e-05, | |
| "loss": 1.1912, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2038735983690112, | |
| "grad_norm": 0.29578036069869995, | |
| "learning_rate": 9.963288406760582e-05, | |
| "loss": 1.0152, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3058103975535168, | |
| "grad_norm": 0.2985854744911194, | |
| "learning_rate": 9.86663298624003e-05, | |
| "loss": 0.957, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4077471967380224, | |
| "grad_norm": 0.2886662185192108, | |
| "learning_rate": 9.710903233782272e-05, | |
| "loss": 0.9539, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.509683995922528, | |
| "grad_norm": 0.29516491293907166, | |
| "learning_rate": 9.497989497625035e-05, | |
| "loss": 0.922, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.6116207951070336, | |
| "grad_norm": 0.2972014844417572, | |
| "learning_rate": 9.230476262104677e-05, | |
| "loss": 0.9182, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7135575942915392, | |
| "grad_norm": 0.31259697675704956, | |
| "learning_rate": 8.911610775517382e-05, | |
| "loss": 0.8974, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.8154943934760448, | |
| "grad_norm": 0.313388854265213, | |
| "learning_rate": 8.545263632923687e-05, | |
| "loss": 0.8888, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.9174311926605505, | |
| "grad_norm": 0.3168061077594757, | |
| "learning_rate": 8.135881792367686e-05, | |
| "loss": 0.8922, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.019367991845056, | |
| "grad_norm": 0.3456219732761383, | |
| "learning_rate": 7.688434594830392e-05, | |
| "loss": 0.8769, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.1213047910295617, | |
| "grad_norm": 0.3286343812942505, | |
| "learning_rate": 7.20835344316187e-05, | |
| "loss": 0.8433, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.2232415902140672, | |
| "grad_norm": 0.40875253081321716, | |
| "learning_rate": 6.701465872208216e-05, | |
| "loss": 0.8328, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.325178389398573, | |
| "grad_norm": 0.38733330368995667, | |
| "learning_rate": 6.173924810432705e-05, | |
| "loss": 0.8472, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.4271151885830786, | |
| "grad_norm": 0.37286508083343506, | |
| "learning_rate": 5.6321338916992315e-05, | |
| "loss": 0.8246, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.529051987767584, | |
| "grad_norm": 0.3836914300918579, | |
| "learning_rate": 5.0826697238317935e-05, | |
| "loss": 0.8198, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.6309887869520896, | |
| "grad_norm": 0.39856258034706116, | |
| "learning_rate": 4.5322020575044114e-05, | |
| "loss": 0.8271, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7329255861365953, | |
| "grad_norm": 0.38490578532218933, | |
| "learning_rate": 3.9874128245030404e-05, | |
| "loss": 0.8203, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.834862385321101, | |
| "grad_norm": 0.4273436367511749, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.815, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.9367991845056065, | |
| "grad_norm": 0.42424914240837097, | |
| "learning_rate": 2.9411724702784758e-05, | |
| "loss": 0.8196, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.038735983690112, | |
| "grad_norm": 0.4148640036582947, | |
| "learning_rate": 2.4524212896808263e-05, | |
| "loss": 0.7939, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.038735983690112, | |
| "eval_loss": 1.0475335121154785, | |
| "eval_runtime": 239.1942, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.140672782874618, | |
| "grad_norm": 0.42366823554039, | |
| "learning_rate": 1.9945942635848748e-05, | |
| "loss": 0.7768, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.2426095820591234, | |
| "grad_norm": 0.4487750232219696, | |
| "learning_rate": 1.5732487918985018e-05, | |
| "loss": 0.7848, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.344546381243629, | |
| "grad_norm": 0.4321724474430084, | |
| "learning_rate": 1.1934994378782772e-05, | |
| "loss": 0.777, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.4464831804281344, | |
| "grad_norm": 0.4505467116832733, | |
| "learning_rate": 8.599558442598998e-06, | |
| "loss": 0.7813, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.5484199796126403, | |
| "grad_norm": 0.4411642849445343, | |
| "learning_rate": 5.766667784397706e-06, | |
| "loss": 0.7694, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.650356778797146, | |
| "grad_norm": 0.44770148396492004, | |
| "learning_rate": 3.470709859234084e-06, | |
| "loss": 0.7721, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.7522935779816513, | |
| "grad_norm": 0.40171849727630615, | |
| "learning_rate": 1.7395544861325718e-06, | |
| "loss": 0.775, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.8542303771661572, | |
| "grad_norm": 0.437456876039505, | |
| "learning_rate": 5.9421554623742e-07, | |
| "loss": 0.7672, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.9561671763506627, | |
| "grad_norm": 0.4368380904197693, | |
| "learning_rate": 4.859590276170556e-08, | |
| "loss": 0.7591, | |
| "step": 1450 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1470, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.452871731469353e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |