| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.008678295582747549, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.678295582747549e-05, | |
| "eval_loss": 1.574290156364441, | |
| "eval_runtime": 661.121, | |
| "eval_samples_per_second": 29.355, | |
| "eval_steps_per_second": 3.67, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002603488674824265, | |
| "grad_norm": 5.837701797485352, | |
| "learning_rate": 1.5e-05, | |
| "loss": 6.4896, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.000520697734964853, | |
| "grad_norm": 5.339621067047119, | |
| "learning_rate": 3e-05, | |
| "loss": 6.5447, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0007810466024472793, | |
| "grad_norm": 5.166632175445557, | |
| "learning_rate": 4.5e-05, | |
| "loss": 6.1585, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0007810466024472793, | |
| "eval_loss": 1.4268044233322144, | |
| "eval_runtime": 666.8349, | |
| "eval_samples_per_second": 29.103, | |
| "eval_steps_per_second": 3.638, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.001041395469929706, | |
| "grad_norm": 3.3273203372955322, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 5.1469, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0013017443374121322, | |
| "grad_norm": 3.0238025188446045, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 4.947, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0015620932048945586, | |
| "grad_norm": 2.8387701511383057, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 4.3529, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0015620932048945586, | |
| "eval_loss": 1.0735480785369873, | |
| "eval_runtime": 667.1413, | |
| "eval_samples_per_second": 29.09, | |
| "eval_steps_per_second": 3.636, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0018224420723769851, | |
| "grad_norm": 3.0103814601898193, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 4.4745, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.002082790939859412, | |
| "grad_norm": 2.6641955375671387, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 3.9389, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.002343139807341838, | |
| "grad_norm": 2.33498215675354, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 3.6797, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.002343139807341838, | |
| "eval_loss": 0.8915460705757141, | |
| "eval_runtime": 667.0273, | |
| "eval_samples_per_second": 29.095, | |
| "eval_steps_per_second": 3.637, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0026034886748242643, | |
| "grad_norm": 2.407578229904175, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 3.656, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.002863837542306691, | |
| "grad_norm": 2.5195510387420654, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 3.4146, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0031241864097891173, | |
| "grad_norm": 2.285335063934326, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 3.0848, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0031241864097891173, | |
| "eval_loss": 0.7998289465904236, | |
| "eval_runtime": 665.3923, | |
| "eval_samples_per_second": 29.166, | |
| "eval_steps_per_second": 3.646, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.003384535277271544, | |
| "grad_norm": 2.3626725673675537, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 3.2387, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0036448841447539702, | |
| "grad_norm": 2.3358616828918457, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 2.9941, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.003905233012236397, | |
| "grad_norm": 2.2276687622070312, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 2.9752, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.003905233012236397, | |
| "eval_loss": 0.7518236041069031, | |
| "eval_runtime": 666.2992, | |
| "eval_samples_per_second": 29.127, | |
| "eval_steps_per_second": 3.641, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.004165581879718824, | |
| "grad_norm": 2.41260027885437, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 2.82, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0044259307472012495, | |
| "grad_norm": 2.4345157146453857, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 2.9344, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.004686279614683676, | |
| "grad_norm": 2.6772620677948, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 2.8991, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.004686279614683676, | |
| "eval_loss": 0.7258293628692627, | |
| "eval_runtime": 667.3649, | |
| "eval_samples_per_second": 29.08, | |
| "eval_steps_per_second": 3.635, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.004946628482166103, | |
| "grad_norm": 2.7180323600769043, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 2.8935, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.005206977349648529, | |
| "grad_norm": 2.3361828327178955, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 2.8249, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.005467326217130955, | |
| "grad_norm": 2.265934467315674, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 2.8911, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.005467326217130955, | |
| "eval_loss": 0.709642767906189, | |
| "eval_runtime": 666.8626, | |
| "eval_samples_per_second": 29.102, | |
| "eval_steps_per_second": 3.638, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.005727675084613382, | |
| "grad_norm": 2.4326114654541016, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 2.9059, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.005988023952095809, | |
| "grad_norm": 2.208724021911621, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 2.7232, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.006248372819578235, | |
| "grad_norm": 2.313538074493408, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 2.79, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.006248372819578235, | |
| "eval_loss": 0.700292706489563, | |
| "eval_runtime": 666.7836, | |
| "eval_samples_per_second": 29.105, | |
| "eval_steps_per_second": 3.638, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.006508721687060661, | |
| "grad_norm": 2.1600091457366943, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 2.7683, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.006769070554543088, | |
| "grad_norm": 2.334808588027954, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 2.7657, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.007029419422025514, | |
| "grad_norm": 2.3936593532562256, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 2.8566, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.007029419422025514, | |
| "eval_loss": 0.6951488256454468, | |
| "eval_runtime": 666.3577, | |
| "eval_samples_per_second": 29.124, | |
| "eval_steps_per_second": 3.641, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.0072897682895079405, | |
| "grad_norm": 2.512799024581909, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 2.7116, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.007550117156990367, | |
| "grad_norm": 2.50602388381958, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 2.7318, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.007810466024472794, | |
| "grad_norm": 2.2158546447753906, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 2.8176, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.007810466024472794, | |
| "eval_loss": 0.6929637789726257, | |
| "eval_runtime": 666.938, | |
| "eval_samples_per_second": 29.099, | |
| "eval_steps_per_second": 3.638, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.00807081489195522, | |
| "grad_norm": 2.417557954788208, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 2.7837, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.008331163759437647, | |
| "grad_norm": 2.59735107421875, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 2.846, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.008591512626920072, | |
| "grad_norm": 2.75762677192688, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 2.8172, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.008591512626920072, | |
| "eval_loss": 0.6925280690193176, | |
| "eval_runtime": 667.4952, | |
| "eval_samples_per_second": 29.074, | |
| "eval_steps_per_second": 3.634, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.2442104332288e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |