| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.04630165528417641, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0004630165528417641, | |
| "eval_loss": 2.0430142879486084, | |
| "eval_runtime": 144.8645, | |
| "eval_samples_per_second": 25.113, | |
| "eval_steps_per_second": 3.141, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0013890496585252923, | |
| "grad_norm": 0.5597144961357117, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.1021, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0027780993170505846, | |
| "grad_norm": 0.5732973217964172, | |
| "learning_rate": 3e-05, | |
| "loss": 1.9359, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004167148975575877, | |
| "grad_norm": 0.6314367055892944, | |
| "learning_rate": 4.5e-05, | |
| "loss": 2.0411, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004167148975575877, | |
| "eval_loss": 2.0033884048461914, | |
| "eval_runtime": 146.5734, | |
| "eval_samples_per_second": 24.82, | |
| "eval_steps_per_second": 3.104, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.005556198634101169, | |
| "grad_norm": 0.7915438413619995, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 2.1273, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0069452482926264615, | |
| "grad_norm": 0.6669667959213257, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 1.8469, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.008334297951151754, | |
| "grad_norm": 0.5791904926300049, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 1.7498, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.008334297951151754, | |
| "eval_loss": 1.7797669172286987, | |
| "eval_runtime": 146.5915, | |
| "eval_samples_per_second": 24.817, | |
| "eval_steps_per_second": 3.104, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.009723347609677045, | |
| "grad_norm": 0.49809256196022034, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 1.8012, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.011112397268202338, | |
| "grad_norm": 0.6324176788330078, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 1.7509, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01250144692672763, | |
| "grad_norm": 0.539840042591095, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 1.7469, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.01250144692672763, | |
| "eval_loss": 1.7146458625793457, | |
| "eval_runtime": 146.5822, | |
| "eval_samples_per_second": 24.819, | |
| "eval_steps_per_second": 3.104, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.013890496585252923, | |
| "grad_norm": 0.47603142261505127, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 1.7088, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.015279546243778214, | |
| "grad_norm": 0.5055053234100342, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 1.753, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.016668595902303508, | |
| "grad_norm": 0.5367181301116943, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 1.6712, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.016668595902303508, | |
| "eval_loss": 1.6824572086334229, | |
| "eval_runtime": 146.4708, | |
| "eval_samples_per_second": 24.838, | |
| "eval_steps_per_second": 3.106, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0180576455608288, | |
| "grad_norm": 0.5550898313522339, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 1.5916, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.01944669521935409, | |
| "grad_norm": 0.5138546824455261, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 1.5893, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.020835744877879384, | |
| "grad_norm": 0.4567825198173523, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 1.6908, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.020835744877879384, | |
| "eval_loss": 1.6608870029449463, | |
| "eval_runtime": 146.5498, | |
| "eval_samples_per_second": 24.824, | |
| "eval_steps_per_second": 3.105, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.022224794536404677, | |
| "grad_norm": 0.5362297892570496, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 1.6418, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.02361384419492997, | |
| "grad_norm": 0.4934613108634949, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 1.6214, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.02500289385345526, | |
| "grad_norm": 0.4781782627105713, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 1.635, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.02500289385345526, | |
| "eval_loss": 1.6491777896881104, | |
| "eval_runtime": 146.6097, | |
| "eval_samples_per_second": 24.814, | |
| "eval_steps_per_second": 3.103, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.026391943511980553, | |
| "grad_norm": 0.449227899312973, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 1.599, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.027780993170505846, | |
| "grad_norm": 0.443278431892395, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 1.4581, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.02917004282903114, | |
| "grad_norm": 0.468054860830307, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 1.6188, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.02917004282903114, | |
| "eval_loss": 1.641588807106018, | |
| "eval_runtime": 146.545, | |
| "eval_samples_per_second": 24.825, | |
| "eval_steps_per_second": 3.105, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.03055909248755643, | |
| "grad_norm": 0.477867066860199, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 1.4842, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.03194814214608172, | |
| "grad_norm": 0.5209053158760071, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 1.6676, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.033337191804607015, | |
| "grad_norm": 0.5418485403060913, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 1.6806, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.033337191804607015, | |
| "eval_loss": 1.6370091438293457, | |
| "eval_runtime": 146.5808, | |
| "eval_samples_per_second": 24.819, | |
| "eval_steps_per_second": 3.104, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.03472624146313231, | |
| "grad_norm": 0.47662776708602905, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 1.6413, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0361152911216576, | |
| "grad_norm": 0.48989754915237427, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 1.6976, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.037504340780182895, | |
| "grad_norm": 0.5410365462303162, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 1.6951, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.037504340780182895, | |
| "eval_loss": 1.6343382596969604, | |
| "eval_runtime": 146.6418, | |
| "eval_samples_per_second": 24.809, | |
| "eval_steps_per_second": 3.103, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.03889339043870818, | |
| "grad_norm": 0.47324180603027344, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 1.6292, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.040282440097233474, | |
| "grad_norm": 0.5699959397315979, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 1.6236, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.04167148975575877, | |
| "grad_norm": 0.49094218015670776, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 1.696, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.04167148975575877, | |
| "eval_loss": 1.6338019371032715, | |
| "eval_runtime": 146.5598, | |
| "eval_samples_per_second": 24.823, | |
| "eval_steps_per_second": 3.105, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.04306053941428406, | |
| "grad_norm": 0.46824368834495544, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 1.6405, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.04444958907280935, | |
| "grad_norm": 0.49031543731689453, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 1.6011, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.04583863873133465, | |
| "grad_norm": 0.48956042528152466, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 1.5216, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.04583863873133465, | |
| "eval_loss": 1.633582353591919, | |
| "eval_runtime": 146.5701, | |
| "eval_samples_per_second": 24.821, | |
| "eval_steps_per_second": 3.104, | |
| "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": 5.58968585846784e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |