| { | |
| "best_metric": 0.9412656309208033, | |
| "best_model_checkpoint": "hubert-large-ls960-ft/checkpoint-12500", | |
| "epoch": 31.998484082870135, | |
| "global_step": 15808, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 9.487666034155598e-06, | |
| "loss": 4.7142, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_accuracy": 0.0, | |
| "eval_f1": 0.0, | |
| "eval_loss": 5.276514053344727, | |
| "eval_precision": 0.0, | |
| "eval_runtime": 207.9912, | |
| "eval_samples_per_second": 12.688, | |
| "eval_steps_per_second": 1.058, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 1.8975332068311197e-05, | |
| "loss": 4.396, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_accuracy": 0.0, | |
| "eval_f1": 0.0, | |
| "eval_loss": 5.414546012878418, | |
| "eval_precision": 0.0, | |
| "eval_runtime": 211.3332, | |
| "eval_samples_per_second": 12.487, | |
| "eval_steps_per_second": 1.041, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 2.846299810246679e-05, | |
| "loss": 3.8883, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "eval_accuracy": 0.0473664266767715, | |
| "eval_f1": 0.010399469379098707, | |
| "eval_loss": 4.433555603027344, | |
| "eval_precision": 0.04084585694141838, | |
| "eval_runtime": 206.03, | |
| "eval_samples_per_second": 12.809, | |
| "eval_steps_per_second": 1.068, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 2.9116468686300697e-05, | |
| "loss": 2.7848, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "eval_accuracy": 0.129973474801061, | |
| "eval_f1": 0.09643252305766986, | |
| "eval_loss": 3.9772207736968994, | |
| "eval_precision": 0.1280919938791079, | |
| "eval_runtime": 212.3184, | |
| "eval_samples_per_second": 12.429, | |
| "eval_steps_per_second": 1.036, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 2.8062135376396992e-05, | |
| "loss": 1.8649, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "eval_accuracy": 0.15763546798029557, | |
| "eval_f1": 0.15466013987384877, | |
| "eval_loss": 3.44816255569458, | |
| "eval_precision": 0.33391187818919965, | |
| "eval_runtime": 206.6675, | |
| "eval_samples_per_second": 12.769, | |
| "eval_steps_per_second": 1.065, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 2.700780206649329e-05, | |
| "loss": 1.3084, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "eval_accuracy": 0.3080712391057219, | |
| "eval_f1": 0.3402390216080642, | |
| "eval_loss": 2.9702537059783936, | |
| "eval_precision": 0.5296031035295795, | |
| "eval_runtime": 212.8576, | |
| "eval_samples_per_second": 12.398, | |
| "eval_steps_per_second": 1.034, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 2.5953468756589585e-05, | |
| "loss": 0.9868, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "eval_accuracy": 0.46873815839333083, | |
| "eval_f1": 0.5353322629682942, | |
| "eval_loss": 2.3984930515289307, | |
| "eval_precision": 0.8031566212997642, | |
| "eval_runtime": 207.1643, | |
| "eval_samples_per_second": 12.739, | |
| "eval_steps_per_second": 1.062, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 2.489913544668588e-05, | |
| "loss": 0.7679, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "eval_accuracy": 0.6521409624857901, | |
| "eval_f1": 0.7095065774545717, | |
| "eval_loss": 1.7936781644821167, | |
| "eval_precision": 0.8388581448735327, | |
| "eval_runtime": 207.3219, | |
| "eval_samples_per_second": 12.729, | |
| "eval_steps_per_second": 1.061, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "learning_rate": 2.3844802136782175e-05, | |
| "loss": 0.6232, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 9.11, | |
| "eval_accuracy": 0.7389162561576355, | |
| "eval_f1": 0.784666736405717, | |
| "eval_loss": 1.4767512083053589, | |
| "eval_precision": 0.8697758947640245, | |
| "eval_runtime": 211.7813, | |
| "eval_samples_per_second": 12.461, | |
| "eval_steps_per_second": 1.039, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "learning_rate": 2.279046882687847e-05, | |
| "loss": 0.5126, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "eval_accuracy": 0.8287230011367942, | |
| "eval_f1": 0.8762583534303519, | |
| "eval_loss": 1.054182529449463, | |
| "eval_precision": 0.9442564310504037, | |
| "eval_runtime": 210.7585, | |
| "eval_samples_per_second": 12.521, | |
| "eval_steps_per_second": 1.044, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "learning_rate": 2.1736135516974768e-05, | |
| "loss": 0.4453, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "eval_accuracy": 0.8518378173550587, | |
| "eval_f1": 0.8959895568662314, | |
| "eval_loss": 0.9049583673477173, | |
| "eval_precision": 0.9511477433978877, | |
| "eval_runtime": 205.4589, | |
| "eval_samples_per_second": 12.844, | |
| "eval_steps_per_second": 1.071, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "learning_rate": 2.0681802207071063e-05, | |
| "loss": 0.3775, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "eval_accuracy": 0.8927624100037893, | |
| "eval_f1": 0.9265578308615539, | |
| "eval_loss": 0.699573278427124, | |
| "eval_precision": 0.9662252548898577, | |
| "eval_runtime": 212.4773, | |
| "eval_samples_per_second": 12.42, | |
| "eval_steps_per_second": 1.035, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "learning_rate": 1.9627468897167357e-05, | |
| "loss": 0.3568, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "eval_accuracy": 0.8957938613111027, | |
| "eval_f1": 0.9284579551378114, | |
| "eval_loss": 0.6156648993492126, | |
| "eval_precision": 0.9743169949637361, | |
| "eval_runtime": 210.2311, | |
| "eval_samples_per_second": 12.553, | |
| "eval_steps_per_second": 1.046, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 14.17, | |
| "learning_rate": 1.8573135587263652e-05, | |
| "loss": 0.3165, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 14.17, | |
| "eval_accuracy": 0.9151193633952255, | |
| "eval_f1": 0.943623023847651, | |
| "eval_loss": 0.4924512505531311, | |
| "eval_precision": 0.9763736094436646, | |
| "eval_runtime": 206.0776, | |
| "eval_samples_per_second": 12.806, | |
| "eval_steps_per_second": 1.068, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 15.18, | |
| "learning_rate": 1.751880227735995e-05, | |
| "loss": 0.2951, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 15.18, | |
| "eval_accuracy": 0.9037514209928003, | |
| "eval_f1": 0.9368584519497015, | |
| "eval_loss": 0.49918055534362793, | |
| "eval_precision": 0.9772932630240506, | |
| "eval_runtime": 205.4956, | |
| "eval_samples_per_second": 12.842, | |
| "eval_steps_per_second": 1.071, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 16.19, | |
| "learning_rate": 1.6464468967456245e-05, | |
| "loss": 0.2763, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 16.19, | |
| "eval_accuracy": 0.9071618037135278, | |
| "eval_f1": 0.9403855453166055, | |
| "eval_loss": 0.5212343335151672, | |
| "eval_precision": 0.9820562398022753, | |
| "eval_runtime": 211.5185, | |
| "eval_samples_per_second": 12.476, | |
| "eval_steps_per_second": 1.04, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 17.21, | |
| "learning_rate": 1.541013565755254e-05, | |
| "loss": 0.2634, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 17.21, | |
| "eval_accuracy": 0.9086775293671845, | |
| "eval_f1": 0.9417778927796274, | |
| "eval_loss": 0.5201326012611389, | |
| "eval_precision": 0.9816838665228488, | |
| "eval_runtime": 205.8774, | |
| "eval_samples_per_second": 12.818, | |
| "eval_steps_per_second": 1.069, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 18.22, | |
| "learning_rate": 1.4355802347648837e-05, | |
| "loss": 0.2422, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 18.22, | |
| "eval_accuracy": 0.9234558544903373, | |
| "eval_f1": 0.9514314684994079, | |
| "eval_loss": 0.45036178827285767, | |
| "eval_precision": 0.9839863223736393, | |
| "eval_runtime": 211.5273, | |
| "eval_samples_per_second": 12.476, | |
| "eval_steps_per_second": 1.04, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "learning_rate": 1.3301469037745133e-05, | |
| "loss": 0.236, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "eval_accuracy": 0.9257294429708223, | |
| "eval_f1": 0.9518288569523149, | |
| "eval_loss": 0.3829096853733063, | |
| "eval_precision": 0.9824861532841103, | |
| "eval_runtime": 210.8744, | |
| "eval_samples_per_second": 12.515, | |
| "eval_steps_per_second": 1.043, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 20.24, | |
| "learning_rate": 1.2247135727841428e-05, | |
| "loss": 0.2272, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 20.24, | |
| "eval_accuracy": 0.9154982948086396, | |
| "eval_f1": 0.9451193658066668, | |
| "eval_loss": 0.4632132947444916, | |
| "eval_precision": 0.9822249030180286, | |
| "eval_runtime": 207.0475, | |
| "eval_samples_per_second": 12.746, | |
| "eval_steps_per_second": 1.063, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "learning_rate": 1.1192802417937724e-05, | |
| "loss": 0.226, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "eval_accuracy": 0.9158772262220538, | |
| "eval_f1": 0.947028332160647, | |
| "eval_loss": 0.47312408685684204, | |
| "eval_precision": 0.9837188228053231, | |
| "eval_runtime": 210.5001, | |
| "eval_samples_per_second": 12.537, | |
| "eval_steps_per_second": 1.045, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 22.27, | |
| "learning_rate": 1.013846910803402e-05, | |
| "loss": 0.2129, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 22.27, | |
| "eval_accuracy": 0.9298976885183782, | |
| "eval_f1": 0.9548846481808738, | |
| "eval_loss": 0.38141778111457825, | |
| "eval_precision": 0.983160805693422, | |
| "eval_runtime": 210.7476, | |
| "eval_samples_per_second": 12.522, | |
| "eval_steps_per_second": 1.044, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 23.28, | |
| "learning_rate": 9.084135798130316e-06, | |
| "loss": 0.2009, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 23.28, | |
| "eval_accuracy": 0.9257294429708223, | |
| "eval_f1": 0.9514840419100091, | |
| "eval_loss": 0.4119464159011841, | |
| "eval_precision": 0.9814366336318854, | |
| "eval_runtime": 206.3284, | |
| "eval_samples_per_second": 12.79, | |
| "eval_steps_per_second": 1.066, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 24.29, | |
| "learning_rate": 8.029802488226612e-06, | |
| "loss": 0.1973, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 24.29, | |
| "eval_accuracy": 0.9215611974232664, | |
| "eval_f1": 0.9492716683321308, | |
| "eval_loss": 0.43100807070732117, | |
| "eval_precision": 0.98427363081955, | |
| "eval_runtime": 210.5932, | |
| "eval_samples_per_second": 12.531, | |
| "eval_steps_per_second": 1.045, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 25.3, | |
| "learning_rate": 6.975469178322908e-06, | |
| "loss": 0.1965, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 25.3, | |
| "eval_accuracy": 0.9412656309208033, | |
| "eval_f1": 0.9627712250405202, | |
| "eval_loss": 0.327240914106369, | |
| "eval_precision": 0.9865248931670977, | |
| "eval_runtime": 207.1296, | |
| "eval_samples_per_second": 12.741, | |
| "eval_steps_per_second": 1.062, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "learning_rate": 5.9211358684192026e-06, | |
| "loss": 0.1989, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "eval_accuracy": 0.9242137173171656, | |
| "eval_f1": 0.9527761591577618, | |
| "eval_loss": 0.4231082797050476, | |
| "eval_precision": 0.9877979282330233, | |
| "eval_runtime": 212.06, | |
| "eval_samples_per_second": 12.445, | |
| "eval_steps_per_second": 1.037, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 27.33, | |
| "learning_rate": 4.866802558515498e-06, | |
| "loss": 0.1916, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 27.33, | |
| "eval_accuracy": 0.9283819628647215, | |
| "eval_f1": 0.9559474626706007, | |
| "eval_loss": 0.3977676033973694, | |
| "eval_precision": 0.9875814057989544, | |
| "eval_runtime": 205.5092, | |
| "eval_samples_per_second": 12.841, | |
| "eval_steps_per_second": 1.071, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 28.34, | |
| "learning_rate": 3.8124692486117947e-06, | |
| "loss": 0.1849, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 28.34, | |
| "eval_accuracy": 0.9215611974232664, | |
| "eval_f1": 0.9506673563440831, | |
| "eval_loss": 0.4528682827949524, | |
| "eval_precision": 0.9865112585967588, | |
| "eval_runtime": 212.4387, | |
| "eval_samples_per_second": 12.422, | |
| "eval_steps_per_second": 1.036, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 29.35, | |
| "learning_rate": 2.7581359387080904e-06, | |
| "loss": 0.1844, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 29.35, | |
| "eval_accuracy": 0.9314134141720348, | |
| "eval_f1": 0.9566213429287339, | |
| "eval_loss": 0.3853737413883209, | |
| "eval_precision": 0.9863541882706378, | |
| "eval_runtime": 205.8351, | |
| "eval_samples_per_second": 12.821, | |
| "eval_steps_per_second": 1.069, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 30.36, | |
| "learning_rate": 1.7038026288043862e-06, | |
| "loss": 0.1831, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 30.36, | |
| "eval_accuracy": 0.9257294429708223, | |
| "eval_f1": 0.9527879636596958, | |
| "eval_loss": 0.41776272654533386, | |
| "eval_precision": 0.9853006909522924, | |
| "eval_runtime": 210.65, | |
| "eval_samples_per_second": 12.528, | |
| "eval_steps_per_second": 1.044, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 31.38, | |
| "learning_rate": 6.494693189006819e-07, | |
| "loss": 0.1778, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 31.38, | |
| "eval_accuracy": 0.9359605911330049, | |
| "eval_f1": 0.9606362253618668, | |
| "eval_loss": 0.37370702624320984, | |
| "eval_precision": 0.9883638648463977, | |
| "eval_runtime": 211.0911, | |
| "eval_samples_per_second": 12.502, | |
| "eval_steps_per_second": 1.042, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "step": 15808, | |
| "total_flos": 1.2696857331997786e+20, | |
| "train_loss": 0.858633176759187, | |
| "train_runtime": 80385.1912, | |
| "train_samples_per_second": 9.454, | |
| "train_steps_per_second": 0.197 | |
| } | |
| ], | |
| "max_steps": 15808, | |
| "num_train_epochs": 32, | |
| "total_flos": 1.2696857331997786e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |