| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0991348744858884, | |
| "eval_steps": 3526, | |
| "global_step": 15500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03545596369309318, | |
| "grad_norm": 2.768805980682373, | |
| "learning_rate": 0.0001964544036306907, | |
| "loss": 3.5372, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.07091192738618636, | |
| "grad_norm": 2.2298574447631836, | |
| "learning_rate": 0.00019290880726138138, | |
| "loss": 2.9277, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.10636789107927953, | |
| "grad_norm": 1.8124895095825195, | |
| "learning_rate": 0.00018936321089207204, | |
| "loss": 2.8216, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.14182385477237272, | |
| "grad_norm": 1.6588631868362427, | |
| "learning_rate": 0.00018581761452276274, | |
| "loss": 2.7462, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.1772798184654659, | |
| "grad_norm": 1.7708145380020142, | |
| "learning_rate": 0.00018227201815345344, | |
| "loss": 2.6757, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.21273578215855907, | |
| "grad_norm": 1.6058692932128906, | |
| "learning_rate": 0.0001787264217841441, | |
| "loss": 2.6299, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.24819174585165224, | |
| "grad_norm": 2.6353378295898438, | |
| "learning_rate": 0.00017518082541483478, | |
| "loss": 2.57, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.2500354559636931, | |
| "eval_loss": 2.3737287521362305, | |
| "eval_runtime": 342.4427, | |
| "eval_samples_per_second": 10.297, | |
| "eval_steps_per_second": 10.297, | |
| "step": 3526 | |
| }, | |
| { | |
| "epoch": 0.28364770954474544, | |
| "grad_norm": 2.034212827682495, | |
| "learning_rate": 0.00017163522904552547, | |
| "loss": 2.5853, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.3191036732378386, | |
| "grad_norm": 3.164950370788574, | |
| "learning_rate": 0.00016808963267621617, | |
| "loss": 2.4973, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.3545596369309318, | |
| "grad_norm": 1.992343783378601, | |
| "learning_rate": 0.0001645440363069068, | |
| "loss": 2.5324, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.39001560062402496, | |
| "grad_norm": 2.540403366088867, | |
| "learning_rate": 0.0001609984399375975, | |
| "loss": 2.4857, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.42547156431711813, | |
| "grad_norm": 5.178898334503174, | |
| "learning_rate": 0.0001574528435682882, | |
| "loss": 2.47, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.4609275280102113, | |
| "grad_norm": 2.17370867729187, | |
| "learning_rate": 0.00015390724719897887, | |
| "loss": 2.4353, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.4963834917033045, | |
| "grad_norm": 2.1408374309539795, | |
| "learning_rate": 0.00015036165082966954, | |
| "loss": 2.4575, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.5000709119273862, | |
| "eval_loss": 2.2161264419555664, | |
| "eval_runtime": 347.4584, | |
| "eval_samples_per_second": 10.148, | |
| "eval_steps_per_second": 10.148, | |
| "step": 7052 | |
| }, | |
| { | |
| "epoch": 0.5318394553963977, | |
| "grad_norm": 2.286954879760742, | |
| "learning_rate": 0.00014681605446036024, | |
| "loss": 2.4082, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.5672954190894909, | |
| "grad_norm": 2.3392558097839355, | |
| "learning_rate": 0.00014327045809105093, | |
| "loss": 2.415, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.602751382782584, | |
| "grad_norm": 2.3896689414978027, | |
| "learning_rate": 0.0001397248617217416, | |
| "loss": 2.3953, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.6382073464756772, | |
| "grad_norm": 2.069913864135742, | |
| "learning_rate": 0.00013617926535243227, | |
| "loss": 2.3731, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.6736633101687703, | |
| "grad_norm": 3.516718864440918, | |
| "learning_rate": 0.00013263366898312297, | |
| "loss": 2.332, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.7091192738618636, | |
| "grad_norm": 2.8783183097839355, | |
| "learning_rate": 0.00012908807261381367, | |
| "loss": 2.2976, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.7445752375549567, | |
| "grad_norm": 3.2221286296844482, | |
| "learning_rate": 0.00012554247624450433, | |
| "loss": 2.2903, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.7501063678910793, | |
| "eval_loss": 2.133788824081421, | |
| "eval_runtime": 340.207, | |
| "eval_samples_per_second": 10.364, | |
| "eval_steps_per_second": 10.364, | |
| "step": 10578 | |
| }, | |
| { | |
| "epoch": 0.7800312012480499, | |
| "grad_norm": 2.0654242038726807, | |
| "learning_rate": 0.00012199687987519502, | |
| "loss": 2.3177, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.8154871649411432, | |
| "grad_norm": 2.5238709449768066, | |
| "learning_rate": 0.0001184512835058857, | |
| "loss": 2.2966, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.8509431286342363, | |
| "grad_norm": 2.6544277667999268, | |
| "learning_rate": 0.00011490568713657637, | |
| "loss": 2.2993, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.8863990923273295, | |
| "grad_norm": 2.4742555618286133, | |
| "learning_rate": 0.00011136009076726705, | |
| "loss": 2.2583, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.9218550560204226, | |
| "grad_norm": 1.410945177078247, | |
| "learning_rate": 0.00010781449439795775, | |
| "loss": 2.2652, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.9573110197135158, | |
| "grad_norm": 2.3635599613189697, | |
| "learning_rate": 0.00010426889802864843, | |
| "loss": 2.3116, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.992766983406609, | |
| "grad_norm": 1.9908702373504639, | |
| "learning_rate": 0.0001007233016593391, | |
| "loss": 2.2715, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.0001418238547723, | |
| "eval_loss": 2.0845015048980713, | |
| "eval_runtime": 347.9194, | |
| "eval_samples_per_second": 10.135, | |
| "eval_steps_per_second": 10.135, | |
| "step": 14104 | |
| }, | |
| { | |
| "epoch": 1.0282229470997022, | |
| "grad_norm": 2.456178903579712, | |
| "learning_rate": 9.717770529002978e-05, | |
| "loss": 2.2921, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.0636789107927953, | |
| "grad_norm": 3.4969823360443115, | |
| "learning_rate": 9.363210892072048e-05, | |
| "loss": 2.216, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.0991348744858884, | |
| "grad_norm": 1.8081468343734741, | |
| "learning_rate": 9.008651255141115e-05, | |
| "loss": 2.242, | |
| "step": 15500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 28204, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.7341146652672e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |