| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9986738753442823, | |
| "eval_steps": 200, | |
| "global_step": 445, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02244210955829848, | |
| "grad_norm": 19.15482521057129, | |
| "learning_rate": 2.222222222222222e-06, | |
| "loss": 3.8476, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04488421911659696, | |
| "grad_norm": 5.446422576904297, | |
| "learning_rate": 4.444444444444444e-06, | |
| "loss": 3.2218, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06732632867489544, | |
| "grad_norm": 1.8523049354553223, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 2.9756, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08976843823319391, | |
| "grad_norm": 2.139192581176758, | |
| "learning_rate": 8.888888888888888e-06, | |
| "loss": 2.88, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11221054779149241, | |
| "grad_norm": 1.853474497795105, | |
| "learning_rate": 9.996145181203616e-06, | |
| "loss": 2.8198, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.13465265734979087, | |
| "grad_norm": 1.501637578010559, | |
| "learning_rate": 9.965342284774633e-06, | |
| "loss": 2.8122, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.15709476690808935, | |
| "grad_norm": 2.0072269439697266, | |
| "learning_rate": 9.903926402016153e-06, | |
| "loss": 2.7806, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.17953687646638783, | |
| "grad_norm": 1.7332258224487305, | |
| "learning_rate": 9.812276182268236e-06, | |
| "loss": 2.7422, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2019789860246863, | |
| "grad_norm": 1.3256088495254517, | |
| "learning_rate": 9.690956679612422e-06, | |
| "loss": 2.736, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.22442109558298481, | |
| "grad_norm": 1.6238477230072021, | |
| "learning_rate": 9.540715869125407e-06, | |
| "loss": 2.7361, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2468632051412833, | |
| "grad_norm": 1.326378583908081, | |
| "learning_rate": 9.362480035363987e-06, | |
| "loss": 2.7135, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.26930531469958174, | |
| "grad_norm": 1.3376497030258179, | |
| "learning_rate": 9.157348061512728e-06, | |
| "loss": 2.7064, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.29174742425788025, | |
| "grad_norm": 1.2815560102462769, | |
| "learning_rate": 8.926584654403725e-06, | |
| "loss": 2.7018, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.3141895338161787, | |
| "grad_norm": 1.5868873596191406, | |
| "learning_rate": 8.671612547178428e-06, | |
| "loss": 2.6961, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.3366316433744772, | |
| "grad_norm": 1.366570234298706, | |
| "learning_rate": 8.39400372766471e-06, | |
| "loss": 2.6968, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.35907375293277566, | |
| "grad_norm": 1.6603009700775146, | |
| "learning_rate": 8.095469746549172e-06, | |
| "loss": 2.6879, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.38151586249107416, | |
| "grad_norm": 1.4688373804092407, | |
| "learning_rate": 7.777851165098012e-06, | |
| "loss": 2.6686, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.4039579720493726, | |
| "grad_norm": 1.2386434078216553, | |
| "learning_rate": 7.443106207484776e-06, | |
| "loss": 2.6497, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.4264000816076711, | |
| "grad_norm": 1.3002716302871704, | |
| "learning_rate": 7.093298687687141e-06, | |
| "loss": 2.6413, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.44884219116596963, | |
| "grad_norm": 1.2603603601455688, | |
| "learning_rate": 6.730585285387465e-06, | |
| "loss": 2.6472, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.44884219116596963, | |
| "eval_loss": 2.642993450164795, | |
| "eval_runtime": 1294.6794, | |
| "eval_samples_per_second": 148.064, | |
| "eval_steps_per_second": 0.842, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4712843007242681, | |
| "grad_norm": 1.2680917978286743, | |
| "learning_rate": 6.3572022493253715e-06, | |
| "loss": 2.6369, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.4937264102825666, | |
| "grad_norm": 1.3160443305969238, | |
| "learning_rate": 5.975451610080643e-06, | |
| "loss": 2.63, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.5161685198408651, | |
| "grad_norm": 1.2467771768569946, | |
| "learning_rate": 5.587686987289189e-06, | |
| "loss": 2.6209, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5386106293991635, | |
| "grad_norm": 1.208018183708191, | |
| "learning_rate": 5.1962990787953436e-06, | |
| "loss": 2.6318, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.561052738957462, | |
| "grad_norm": 1.2416397333145142, | |
| "learning_rate": 4.803700921204659e-06, | |
| "loss": 2.621, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5834948485157605, | |
| "grad_norm": 1.1826361417770386, | |
| "learning_rate": 4.4123130127108125e-06, | |
| "loss": 2.6161, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.605936958074059, | |
| "grad_norm": 1.2550407648086548, | |
| "learning_rate": 4.02454838991936e-06, | |
| "loss": 2.6164, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.6283790676323574, | |
| "grad_norm": 1.2681384086608887, | |
| "learning_rate": 3.6427977506746293e-06, | |
| "loss": 2.6091, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.6508211771906559, | |
| "grad_norm": 1.2637056112289429, | |
| "learning_rate": 3.269414714612534e-06, | |
| "loss": 2.5967, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.6732632867489544, | |
| "grad_norm": 1.211774468421936, | |
| "learning_rate": 2.906701312312861e-06, | |
| "loss": 2.6031, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6957053963072529, | |
| "grad_norm": 1.1411036252975464, | |
| "learning_rate": 2.5568937925152272e-06, | |
| "loss": 2.6014, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.7181475058655513, | |
| "grad_norm": 1.1422080993652344, | |
| "learning_rate": 2.2221488349019903e-06, | |
| "loss": 2.5978, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.7405896154238498, | |
| "grad_norm": 1.172059416770935, | |
| "learning_rate": 1.9045302534508298e-06, | |
| "loss": 2.5911, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.7630317249821483, | |
| "grad_norm": 1.1655080318450928, | |
| "learning_rate": 1.6059962723352912e-06, | |
| "loss": 2.5913, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.7854738345404468, | |
| "grad_norm": 1.1286932229995728, | |
| "learning_rate": 1.3283874528215735e-06, | |
| "loss": 2.5819, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.8079159440987452, | |
| "grad_norm": 1.1322216987609863, | |
| "learning_rate": 1.0734153455962765e-06, | |
| "loss": 2.5833, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.8303580536570437, | |
| "grad_norm": 1.1392606496810913, | |
| "learning_rate": 8.426519384872733e-07, | |
| "loss": 2.5851, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.8528001632153422, | |
| "grad_norm": 1.1811796426773071, | |
| "learning_rate": 6.375199646360142e-07, | |
| "loss": 2.5853, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.8752422727736408, | |
| "grad_norm": 1.1267277002334595, | |
| "learning_rate": 4.5928413087459325e-07, | |
| "loss": 2.5832, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.8976843823319393, | |
| "grad_norm": 1.116821527481079, | |
| "learning_rate": 3.0904332038757977e-07, | |
| "loss": 2.5779, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8976843823319393, | |
| "eval_loss": 2.582942485809326, | |
| "eval_runtime": 1274.0632, | |
| "eval_samples_per_second": 150.46, | |
| "eval_steps_per_second": 0.856, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.9201264918902377, | |
| "grad_norm": 1.1507278680801392, | |
| "learning_rate": 1.8772381773176417e-07, | |
| "loss": 2.5833, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.9425686014485362, | |
| "grad_norm": 1.0935174226760864, | |
| "learning_rate": 9.607359798384785e-08, | |
| "loss": 2.597, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9650107110068347, | |
| "grad_norm": 1.1115341186523438, | |
| "learning_rate": 3.465771522536854e-08, | |
| "loss": 2.5725, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.9874528205651332, | |
| "grad_norm": 1.131402611732483, | |
| "learning_rate": 3.854818796385495e-09, | |
| "loss": 2.575, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.9986738753442823, | |
| "step": 445, | |
| "total_flos": 7.860958022007259e+18, | |
| "train_loss": 2.698132219207421, | |
| "train_runtime": 39941.2631, | |
| "train_samples_per_second": 43.195, | |
| "train_steps_per_second": 0.011 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 445, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.860958022007259e+18, | |
| "train_batch_size": 22, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |