| { | |
| "best_metric": 1.9448524713516235, | |
| "best_model_checkpoint": "miner_id_24/checkpoint-500", | |
| "epoch": 0.7256894049346879, | |
| "eval_steps": 50, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001451378809869376, | |
| "eval_loss": 2.528571605682373, | |
| "eval_runtime": 24.5264, | |
| "eval_samples_per_second": 11.824, | |
| "eval_steps_per_second": 2.976, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01451378809869376, | |
| "grad_norm": 1.4658918380737305, | |
| "learning_rate": 4.2600000000000005e-05, | |
| "loss": 2.3421, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02902757619738752, | |
| "grad_norm": 1.2775057554244995, | |
| "learning_rate": 8.520000000000001e-05, | |
| "loss": 2.1831, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04354136429608128, | |
| "grad_norm": 1.5997238159179688, | |
| "learning_rate": 0.0001278, | |
| "loss": 2.2168, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05805515239477504, | |
| "grad_norm": 1.8565945625305176, | |
| "learning_rate": 0.00017040000000000002, | |
| "loss": 2.2017, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07256894049346879, | |
| "grad_norm": 5.11404275894165, | |
| "learning_rate": 0.000213, | |
| "loss": 2.2357, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07256894049346879, | |
| "eval_loss": 2.3870110511779785, | |
| "eval_runtime": 24.5991, | |
| "eval_samples_per_second": 11.789, | |
| "eval_steps_per_second": 2.968, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08708272859216255, | |
| "grad_norm": 1.203863263130188, | |
| "learning_rate": 0.00021274057135267128, | |
| "loss": 2.2284, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.10159651669085631, | |
| "grad_norm": 1.2065693140029907, | |
| "learning_rate": 0.00021196354932097723, | |
| "loss": 2.1552, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.11611030478955008, | |
| "grad_norm": 1.6284523010253906, | |
| "learning_rate": 0.0002106727194781503, | |
| "loss": 2.181, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.13062409288824384, | |
| "grad_norm": 1.8947309255599976, | |
| "learning_rate": 0.00020887437061743096, | |
| "loss": 2.1743, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.14513788098693758, | |
| "grad_norm": 3.7555911540985107, | |
| "learning_rate": 0.00020657726411369925, | |
| "loss": 2.1284, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.14513788098693758, | |
| "eval_loss": 2.3372433185577393, | |
| "eval_runtime": 24.4036, | |
| "eval_samples_per_second": 11.884, | |
| "eval_steps_per_second": 2.991, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15965166908563136, | |
| "grad_norm": 1.1661285161972046, | |
| "learning_rate": 0.000203792591238937, | |
| "loss": 2.1987, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.1741654571843251, | |
| "grad_norm": 1.179563045501709, | |
| "learning_rate": 0.0002005339186394757, | |
| "loss": 2.1252, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.18867924528301888, | |
| "grad_norm": 1.3845571279525757, | |
| "learning_rate": 0.00019681712224065936, | |
| "loss": 2.1114, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.20319303338171263, | |
| "grad_norm": 1.61588454246521, | |
| "learning_rate": 0.0001926603099009319, | |
| "loss": 2.1507, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.21770682148040638, | |
| "grad_norm": 3.884338140487671, | |
| "learning_rate": 0.00018808373319217114, | |
| "loss": 2.0528, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.21770682148040638, | |
| "eval_loss": 2.194847345352173, | |
| "eval_runtime": 24.5982, | |
| "eval_samples_per_second": 11.789, | |
| "eval_steps_per_second": 2.968, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.23222060957910015, | |
| "grad_norm": 1.1318994760513306, | |
| "learning_rate": 0.00018310968873606635, | |
| "loss": 2.1511, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.2467343976777939, | |
| "grad_norm": 1.1854448318481445, | |
| "learning_rate": 0.0001777624095772184, | |
| "loss": 2.0839, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.2612481857764877, | |
| "grad_norm": 1.4225796461105347, | |
| "learning_rate": 0.0001720679471221826, | |
| "loss": 2.0992, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.2757619738751814, | |
| "grad_norm": 1.514849305152893, | |
| "learning_rate": 0.00016605404421963453, | |
| "loss": 2.1068, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.29027576197387517, | |
| "grad_norm": 4.458597660064697, | |
| "learning_rate": 0.00015975, | |
| "loss": 2.0656, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.29027576197387517, | |
| "eval_loss": 2.1229233741760254, | |
| "eval_runtime": 24.8076, | |
| "eval_samples_per_second": 11.69, | |
| "eval_steps_per_second": 2.943, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3047895500725689, | |
| "grad_norm": 1.1364762783050537, | |
| "learning_rate": 0.00015318652713303674, | |
| "loss": 2.144, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.3193033381712627, | |
| "grad_norm": 1.0806138515472412, | |
| "learning_rate": 0.00014639560219879464, | |
| "loss": 2.0368, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.33381712626995647, | |
| "grad_norm": 1.2863917350769043, | |
| "learning_rate": 0.0001394103099009319, | |
| "loss": 2.0586, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.3483309143686502, | |
| "grad_norm": 1.7007073163986206, | |
| "learning_rate": 0.0001322646818813646, | |
| "loss": 2.058, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.36284470246734396, | |
| "grad_norm": 2.966454029083252, | |
| "learning_rate": 0.0001249935309215281, | |
| "loss": 2.0311, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.36284470246734396, | |
| "eval_loss": 2.0804994106292725, | |
| "eval_runtime": 24.4819, | |
| "eval_samples_per_second": 11.845, | |
| "eval_steps_per_second": 2.982, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.37735849056603776, | |
| "grad_norm": 1.0515992641448975, | |
| "learning_rate": 0.0001176322813380051, | |
| "loss": 2.0973, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.3918722786647315, | |
| "grad_norm": 1.0517977476119995, | |
| "learning_rate": 0.00011021679639881638, | |
| "loss": 1.9887, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.40638606676342526, | |
| "grad_norm": 1.2678823471069336, | |
| "learning_rate": 0.00010278320360118368, | |
| "loss": 2.0417, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.420899854862119, | |
| "grad_norm": 1.3994240760803223, | |
| "learning_rate": 9.536771866199493e-05, | |
| "loss": 1.9758, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.43541364296081275, | |
| "grad_norm": 3.492399215698242, | |
| "learning_rate": 8.800646907847192e-05, | |
| "loss": 1.9881, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.43541364296081275, | |
| "eval_loss": 2.0401976108551025, | |
| "eval_runtime": 24.604, | |
| "eval_samples_per_second": 11.787, | |
| "eval_steps_per_second": 2.967, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.44992743105950656, | |
| "grad_norm": 0.9537028074264526, | |
| "learning_rate": 8.07353181186354e-05, | |
| "loss": 2.0359, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.4644412191582003, | |
| "grad_norm": 1.1064801216125488, | |
| "learning_rate": 7.35896900990681e-05, | |
| "loss": 1.9925, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.47895500725689405, | |
| "grad_norm": 1.2894712686538696, | |
| "learning_rate": 6.660439780120536e-05, | |
| "loss": 2.0071, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.4934687953555878, | |
| "grad_norm": 1.4024039506912231, | |
| "learning_rate": 5.981347286696324e-05, | |
| "loss": 2.0153, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.5079825834542816, | |
| "grad_norm": 3.1581623554229736, | |
| "learning_rate": 5.325000000000002e-05, | |
| "loss": 1.9632, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.5079825834542816, | |
| "eval_loss": 1.9959750175476074, | |
| "eval_runtime": 24.4071, | |
| "eval_samples_per_second": 11.882, | |
| "eval_steps_per_second": 2.991, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.5224963715529753, | |
| "grad_norm": 1.0032169818878174, | |
| "learning_rate": 4.6945955780365475e-05, | |
| "loss": 2.0021, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.5370101596516691, | |
| "grad_norm": 0.9521530866622925, | |
| "learning_rate": 4.0932052877817393e-05, | |
| "loss": 1.9331, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.5515239477503628, | |
| "grad_norm": 1.1082749366760254, | |
| "learning_rate": 3.523759042278163e-05, | |
| "loss": 1.9542, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.5660377358490566, | |
| "grad_norm": 1.272968053817749, | |
| "learning_rate": 2.989031126393367e-05, | |
| "loss": 1.9261, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.5805515239477503, | |
| "grad_norm": 2.808889865875244, | |
| "learning_rate": 2.4916266807828855e-05, | |
| "loss": 1.8051, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5805515239477503, | |
| "eval_loss": 1.9632947444915771, | |
| "eval_runtime": 24.3908, | |
| "eval_samples_per_second": 11.89, | |
| "eval_steps_per_second": 2.993, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5950653120464441, | |
| "grad_norm": 0.8925508856773376, | |
| "learning_rate": 2.033969009906811e-05, | |
| "loss": 1.9807, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.6095791001451378, | |
| "grad_norm": 0.901614248752594, | |
| "learning_rate": 1.6182877759340637e-05, | |
| "loss": 1.9157, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.6240928882438317, | |
| "grad_norm": 1.059577226638794, | |
| "learning_rate": 1.2466081360524275e-05, | |
| "loss": 1.9516, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.6386066763425254, | |
| "grad_norm": 1.138191819190979, | |
| "learning_rate": 9.207408761062996e-06, | |
| "loss": 1.9491, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.6531204644412192, | |
| "grad_norm": 3.2761425971984863, | |
| "learning_rate": 6.422735886300764e-06, | |
| "loss": 1.889, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.6531204644412192, | |
| "eval_loss": 1.9480090141296387, | |
| "eval_runtime": 24.632, | |
| "eval_samples_per_second": 11.773, | |
| "eval_steps_per_second": 2.964, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.6676342525399129, | |
| "grad_norm": 0.9090232253074646, | |
| "learning_rate": 4.125629382569038e-06, | |
| "loss": 1.9756, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.6821480406386067, | |
| "grad_norm": 0.8540071249008179, | |
| "learning_rate": 2.327280521849694e-06, | |
| "loss": 1.9429, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.6966618287373004, | |
| "grad_norm": 1.0961964130401611, | |
| "learning_rate": 1.0364506790227565e-06, | |
| "loss": 1.9158, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.7111756168359942, | |
| "grad_norm": 1.251514196395874, | |
| "learning_rate": 2.5942864732872295e-07, | |
| "loss": 1.9717, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.7256894049346879, | |
| "grad_norm": 2.7084903717041016, | |
| "learning_rate": 0.0, | |
| "loss": 1.8717, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7256894049346879, | |
| "eval_loss": 1.9448524713516235, | |
| "eval_runtime": 24.6136, | |
| "eval_samples_per_second": 11.782, | |
| "eval_steps_per_second": 2.966, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2944932147612877e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |