| { | |
| "best_metric": 0.2951599955558777, | |
| "best_model_checkpoint": "/home/datawork-iot-nos/Seatizen/models/multilabel/fine_scale/drone_DinoVdeau-large-2024_09_18-batch-size64_epochs100_freeze/checkpoint-11584", | |
| "epoch": 74.0, | |
| "eval_steps": 500, | |
| "global_step": 13394, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.240894901144641, | |
| "eval_f1_macro": 0.5745037518359608, | |
| "eval_f1_micro": 0.8203088441740758, | |
| "eval_loss": 0.33405476808547974, | |
| "eval_roc_auc": 0.8507572583590943, | |
| "eval_runtime": 70.5588, | |
| "eval_samples_per_second": 54.479, | |
| "eval_steps_per_second": 0.865, | |
| "learning_rate": 0.001, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.24011446409989595, | |
| "eval_f1_macro": 0.5964117272843149, | |
| "eval_f1_micro": 0.8326538829332126, | |
| "eval_loss": 0.31857866048812866, | |
| "eval_roc_auc": 0.861811323650532, | |
| "eval_runtime": 79.7549, | |
| "eval_samples_per_second": 48.198, | |
| "eval_steps_per_second": 0.765, | |
| "learning_rate": 0.001, | |
| "step": 362 | |
| }, | |
| { | |
| "epoch": 2.7624309392265194, | |
| "grad_norm": 0.4326592683792114, | |
| "learning_rate": 0.001, | |
| "loss": 0.41, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.2515608740894901, | |
| "eval_f1_macro": 0.5896466586231784, | |
| "eval_f1_micro": 0.8322937087863127, | |
| "eval_loss": 0.3189840316772461, | |
| "eval_roc_auc": 0.8611107288488111, | |
| "eval_runtime": 73.8912, | |
| "eval_samples_per_second": 52.022, | |
| "eval_steps_per_second": 0.826, | |
| "learning_rate": 0.001, | |
| "step": 543 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.2572840790842872, | |
| "eval_f1_macro": 0.5800422752857416, | |
| "eval_f1_micro": 0.8322264673453166, | |
| "eval_loss": 0.31490686535835266, | |
| "eval_roc_auc": 0.8608139926082374, | |
| "eval_runtime": 74.4071, | |
| "eval_samples_per_second": 51.662, | |
| "eval_steps_per_second": 0.82, | |
| "learning_rate": 0.001, | |
| "step": 724 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.25260145681581686, | |
| "eval_f1_macro": 0.5992015270576083, | |
| "eval_f1_micro": 0.831389835721059, | |
| "eval_loss": 0.3153345584869385, | |
| "eval_roc_auc": 0.860148186700414, | |
| "eval_runtime": 75.6696, | |
| "eval_samples_per_second": 50.8, | |
| "eval_steps_per_second": 0.806, | |
| "learning_rate": 0.001, | |
| "step": 905 | |
| }, | |
| { | |
| "epoch": 5.524861878453039, | |
| "grad_norm": 0.3112102448940277, | |
| "learning_rate": 0.001, | |
| "loss": 0.3412, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.2505202913631634, | |
| "eval_f1_macro": 0.6058921391497434, | |
| "eval_f1_micro": 0.8368403933244926, | |
| "eval_loss": 0.3146817088127136, | |
| "eval_roc_auc": 0.8655139600288801, | |
| "eval_runtime": 71.4939, | |
| "eval_samples_per_second": 53.767, | |
| "eval_steps_per_second": 0.853, | |
| "learning_rate": 0.001, | |
| "step": 1086 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.2643080124869927, | |
| "eval_f1_macro": 0.5671643866070547, | |
| "eval_f1_micro": 0.8297352130260625, | |
| "eval_loss": 0.31298699975013733, | |
| "eval_roc_auc": 0.858329231354068, | |
| "eval_runtime": 72.5707, | |
| "eval_samples_per_second": 52.969, | |
| "eval_steps_per_second": 0.841, | |
| "learning_rate": 0.001, | |
| "step": 1267 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.2643080124869927, | |
| "eval_f1_macro": 0.5794085884877922, | |
| "eval_f1_micro": 0.8304935064935066, | |
| "eval_loss": 0.31266748905181885, | |
| "eval_roc_auc": 0.8589842911078356, | |
| "eval_runtime": 68.868, | |
| "eval_samples_per_second": 55.817, | |
| "eval_steps_per_second": 0.886, | |
| "learning_rate": 0.001, | |
| "step": 1448 | |
| }, | |
| { | |
| "epoch": 8.287292817679559, | |
| "grad_norm": 0.2548049986362457, | |
| "learning_rate": 0.001, | |
| "loss": 0.3338, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.2700312174817898, | |
| "eval_f1_macro": 0.5719440884615888, | |
| "eval_f1_micro": 0.8326600372902424, | |
| "eval_loss": 0.31311556696891785, | |
| "eval_roc_auc": 0.8608317149678372, | |
| "eval_runtime": 68.5911, | |
| "eval_samples_per_second": 56.042, | |
| "eval_steps_per_second": 0.889, | |
| "learning_rate": 0.001, | |
| "step": 1629 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.2518210197710718, | |
| "eval_f1_macro": 0.5897148649683502, | |
| "eval_f1_micro": 0.8355268182165314, | |
| "eval_loss": 0.3096640110015869, | |
| "eval_roc_auc": 0.8637739072507734, | |
| "eval_runtime": 68.7801, | |
| "eval_samples_per_second": 55.888, | |
| "eval_steps_per_second": 0.887, | |
| "learning_rate": 0.001, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.27419354838709675, | |
| "eval_f1_macro": 0.5735365394193856, | |
| "eval_f1_micro": 0.8332294264339152, | |
| "eval_loss": 0.31233683228492737, | |
| "eval_roc_auc": 0.8611851168597349, | |
| "eval_runtime": 66.8379, | |
| "eval_samples_per_second": 57.512, | |
| "eval_steps_per_second": 0.913, | |
| "learning_rate": 0.001, | |
| "step": 1991 | |
| }, | |
| { | |
| "epoch": 11.049723756906078, | |
| "grad_norm": 0.24715092778205872, | |
| "learning_rate": 0.001, | |
| "loss": 0.3303, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.27055150884495316, | |
| "eval_f1_macro": 0.5805560421457799, | |
| "eval_f1_micro": 0.8331433309151208, | |
| "eval_loss": 0.3085036873817444, | |
| "eval_roc_auc": 0.8612045319689331, | |
| "eval_runtime": 67.4806, | |
| "eval_samples_per_second": 56.965, | |
| "eval_steps_per_second": 0.904, | |
| "learning_rate": 0.001, | |
| "step": 2172 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.2645681581685744, | |
| "eval_f1_macro": 0.5951616619649476, | |
| "eval_f1_micro": 0.8347951336196919, | |
| "eval_loss": 0.3079213500022888, | |
| "eval_roc_auc": 0.8627997612661887, | |
| "eval_runtime": 64.7469, | |
| "eval_samples_per_second": 59.37, | |
| "eval_steps_per_second": 0.942, | |
| "learning_rate": 0.001, | |
| "step": 2353 | |
| }, | |
| { | |
| "epoch": 13.812154696132596, | |
| "grad_norm": 0.2406352013349533, | |
| "learning_rate": 0.001, | |
| "loss": 0.3278, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.2533818938605619, | |
| "eval_f1_macro": 0.5969453701098729, | |
| "eval_f1_micro": 0.8339762460663892, | |
| "eval_loss": 0.31647807359695435, | |
| "eval_roc_auc": 0.8626087016827515, | |
| "eval_runtime": 65.4843, | |
| "eval_samples_per_second": 58.701, | |
| "eval_steps_per_second": 0.932, | |
| "learning_rate": 0.001, | |
| "step": 2534 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.27601456815816855, | |
| "eval_f1_macro": 0.5790027839889411, | |
| "eval_f1_micro": 0.8351292272084352, | |
| "eval_loss": 0.3074161410331726, | |
| "eval_roc_auc": 0.8630771728562265, | |
| "eval_runtime": 64.7162, | |
| "eval_samples_per_second": 59.398, | |
| "eval_steps_per_second": 0.943, | |
| "learning_rate": 0.001, | |
| "step": 2715 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.26352757544224764, | |
| "eval_f1_macro": 0.5888561594424128, | |
| "eval_f1_micro": 0.8355780022446689, | |
| "eval_loss": 0.3094619810581207, | |
| "eval_roc_auc": 0.8637418270912655, | |
| "eval_runtime": 70.1431, | |
| "eval_samples_per_second": 54.802, | |
| "eval_steps_per_second": 0.87, | |
| "learning_rate": 0.001, | |
| "step": 2896 | |
| }, | |
| { | |
| "epoch": 16.574585635359117, | |
| "grad_norm": 0.1948590725660324, | |
| "learning_rate": 0.001, | |
| "loss": 0.3273, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.24739854318418314, | |
| "eval_f1_macro": 0.6137563857640569, | |
| "eval_f1_micro": 0.8395418326693228, | |
| "eval_loss": 0.31031784415245056, | |
| "eval_roc_auc": 0.8679706824480403, | |
| "eval_runtime": 71.0586, | |
| "eval_samples_per_second": 54.096, | |
| "eval_steps_per_second": 0.858, | |
| "learning_rate": 0.001, | |
| "step": 3077 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.2762747138397503, | |
| "eval_f1_macro": 0.5714548681788686, | |
| "eval_f1_micro": 0.8334075140725227, | |
| "eval_loss": 0.3063325881958008, | |
| "eval_roc_auc": 0.8610597084880822, | |
| "eval_runtime": 70.7944, | |
| "eval_samples_per_second": 54.298, | |
| "eval_steps_per_second": 0.862, | |
| "learning_rate": 0.001, | |
| "step": 3258 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.26560874089490116, | |
| "eval_f1_macro": 0.5919936165947856, | |
| "eval_f1_micro": 0.8336647470495572, | |
| "eval_loss": 0.3109950125217438, | |
| "eval_roc_auc": 0.8617398269664828, | |
| "eval_runtime": 71.2213, | |
| "eval_samples_per_second": 53.973, | |
| "eval_steps_per_second": 0.856, | |
| "learning_rate": 0.001, | |
| "step": 3439 | |
| }, | |
| { | |
| "epoch": 19.337016574585636, | |
| "grad_norm": 0.19638165831565857, | |
| "learning_rate": 0.001, | |
| "loss": 0.324, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.2596253902185224, | |
| "eval_f1_macro": 0.5983991279635611, | |
| "eval_f1_micro": 0.8375498995652063, | |
| "eval_loss": 0.3072282373905182, | |
| "eval_roc_auc": 0.8654821738600246, | |
| "eval_runtime": 69.3213, | |
| "eval_samples_per_second": 55.452, | |
| "eval_steps_per_second": 0.88, | |
| "learning_rate": 0.001, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.2559833506763788, | |
| "eval_f1_macro": 0.6089994809825815, | |
| "eval_f1_micro": 0.8388680190333083, | |
| "eval_loss": 0.30741065740585327, | |
| "eval_roc_auc": 0.8671726276510136, | |
| "eval_runtime": 71.9913, | |
| "eval_samples_per_second": 53.395, | |
| "eval_steps_per_second": 0.847, | |
| "learning_rate": 0.001, | |
| "step": 3801 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.26560874089490116, | |
| "eval_f1_macro": 0.580833002136296, | |
| "eval_f1_micro": 0.8355413783345766, | |
| "eval_loss": 0.3070025146007538, | |
| "eval_roc_auc": 0.8634013163963336, | |
| "eval_runtime": 70.2825, | |
| "eval_samples_per_second": 54.694, | |
| "eval_steps_per_second": 0.868, | |
| "learning_rate": 0.001, | |
| "step": 3982 | |
| }, | |
| { | |
| "epoch": 22.099447513812155, | |
| "grad_norm": 0.21378174424171448, | |
| "learning_rate": 0.001, | |
| "loss": 0.3263, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.2627471383975026, | |
| "eval_f1_macro": 0.6160376071910248, | |
| "eval_f1_micro": 0.8388706831358709, | |
| "eval_loss": 0.3077291250228882, | |
| "eval_roc_auc": 0.8669128705765787, | |
| "eval_runtime": 71.2675, | |
| "eval_samples_per_second": 53.938, | |
| "eval_steps_per_second": 0.856, | |
| "learning_rate": 0.001, | |
| "step": 4163 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.26560874089490116, | |
| "eval_f1_macro": 0.5881215193388629, | |
| "eval_f1_micro": 0.8362589632218766, | |
| "eval_loss": 0.30613505840301514, | |
| "eval_roc_auc": 0.8640193969839625, | |
| "eval_runtime": 71.2696, | |
| "eval_samples_per_second": 53.936, | |
| "eval_steps_per_second": 0.856, | |
| "learning_rate": 0.001, | |
| "step": 4344 | |
| }, | |
| { | |
| "epoch": 24.861878453038674, | |
| "grad_norm": 0.1702233850955963, | |
| "learning_rate": 0.001, | |
| "loss": 0.3244, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.2663891779396462, | |
| "eval_f1_macro": 0.6102297364067529, | |
| "eval_f1_micro": 0.8402073587052726, | |
| "eval_loss": 0.3043115735054016, | |
| "eval_roc_auc": 0.8678931350845757, | |
| "eval_runtime": 75.5305, | |
| "eval_samples_per_second": 50.893, | |
| "eval_steps_per_second": 0.808, | |
| "learning_rate": 0.001, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.26586888657648283, | |
| "eval_f1_macro": 0.5805627478330566, | |
| "eval_f1_micro": 0.8326726046439789, | |
| "eval_loss": 0.31102412939071655, | |
| "eval_roc_auc": 0.8610437919514087, | |
| "eval_runtime": 75.2782, | |
| "eval_samples_per_second": 51.064, | |
| "eval_steps_per_second": 0.81, | |
| "learning_rate": 0.001, | |
| "step": 4706 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.2713319458896982, | |
| "eval_f1_macro": 0.585029632368333, | |
| "eval_f1_micro": 0.8379808306709265, | |
| "eval_loss": 0.30523133277893066, | |
| "eval_roc_auc": 0.8656351026045385, | |
| "eval_runtime": 73.6801, | |
| "eval_samples_per_second": 52.171, | |
| "eval_steps_per_second": 0.828, | |
| "learning_rate": 0.001, | |
| "step": 4887 | |
| }, | |
| { | |
| "epoch": 27.624309392265193, | |
| "grad_norm": 0.15147489309310913, | |
| "learning_rate": 0.001, | |
| "loss": 0.3257, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.2663891779396462, | |
| "eval_f1_macro": 0.5973785723470736, | |
| "eval_f1_micro": 0.8397143725311454, | |
| "eval_loss": 0.3030068874359131, | |
| "eval_roc_auc": 0.8674308050491781, | |
| "eval_runtime": 74.1975, | |
| "eval_samples_per_second": 51.808, | |
| "eval_steps_per_second": 0.822, | |
| "learning_rate": 0.001, | |
| "step": 5068 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.2666493236212279, | |
| "eval_f1_macro": 0.5901889833654033, | |
| "eval_f1_micro": 0.8362408553742262, | |
| "eval_loss": 0.30669936537742615, | |
| "eval_roc_auc": 0.8641808816102908, | |
| "eval_runtime": 71.9481, | |
| "eval_samples_per_second": 53.427, | |
| "eval_steps_per_second": 0.848, | |
| "learning_rate": 0.001, | |
| "step": 5249 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.26352757544224764, | |
| "eval_f1_macro": 0.5923557708061385, | |
| "eval_f1_micro": 0.83628025477707, | |
| "eval_loss": 0.3061116933822632, | |
| "eval_roc_auc": 0.8643618491898142, | |
| "eval_runtime": 77.3574, | |
| "eval_samples_per_second": 49.691, | |
| "eval_steps_per_second": 0.789, | |
| "learning_rate": 0.001, | |
| "step": 5430 | |
| }, | |
| { | |
| "epoch": 30.386740331491712, | |
| "grad_norm": 0.1558382362127304, | |
| "learning_rate": 0.001, | |
| "loss": 0.3243, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.2708116545265349, | |
| "eval_f1_macro": 0.5866580729382617, | |
| "eval_f1_micro": 0.8373469177729752, | |
| "eval_loss": 0.30276864767074585, | |
| "eval_roc_auc": 0.8648652901745116, | |
| "eval_runtime": 67.6643, | |
| "eval_samples_per_second": 56.81, | |
| "eval_steps_per_second": 0.902, | |
| "learning_rate": 0.001, | |
| "step": 5611 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.25676378772112385, | |
| "eval_f1_macro": 0.60937867416881, | |
| "eval_f1_micro": 0.8387781935875391, | |
| "eval_loss": 0.30601420998573303, | |
| "eval_roc_auc": 0.8667133907993204, | |
| "eval_runtime": 71.6477, | |
| "eval_samples_per_second": 53.651, | |
| "eval_steps_per_second": 0.851, | |
| "learning_rate": 0.001, | |
| "step": 5792 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.26508844953173777, | |
| "eval_f1_macro": 0.586599207562497, | |
| "eval_f1_micro": 0.8341955642063208, | |
| "eval_loss": 0.30689385533332825, | |
| "eval_roc_auc": 0.8624638555872349, | |
| "eval_runtime": 74.1949, | |
| "eval_samples_per_second": 51.809, | |
| "eval_steps_per_second": 0.822, | |
| "learning_rate": 0.001, | |
| "step": 5973 | |
| }, | |
| { | |
| "epoch": 33.149171270718234, | |
| "grad_norm": 0.1734226495027542, | |
| "learning_rate": 0.001, | |
| "loss": 0.3257, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.2663891779396462, | |
| "eval_f1_macro": 0.5901389994189593, | |
| "eval_f1_micro": 0.8362871579163353, | |
| "eval_loss": 0.3069196939468384, | |
| "eval_roc_auc": 0.8641186401875288, | |
| "eval_runtime": 74.4957, | |
| "eval_samples_per_second": 51.6, | |
| "eval_steps_per_second": 0.819, | |
| "learning_rate": 0.001, | |
| "step": 6154 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.2627471383975026, | |
| "eval_f1_macro": 0.6009387049254514, | |
| "eval_f1_micro": 0.8379563532531105, | |
| "eval_loss": 0.30412742495536804, | |
| "eval_roc_auc": 0.8657129213439902, | |
| "eval_runtime": 67.827, | |
| "eval_samples_per_second": 56.674, | |
| "eval_steps_per_second": 0.899, | |
| "learning_rate": 0.001, | |
| "step": 6335 | |
| }, | |
| { | |
| "epoch": 35.91160220994475, | |
| "grad_norm": 0.14143767952919006, | |
| "learning_rate": 0.001, | |
| "loss": 0.324, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.2661290322580645, | |
| "eval_f1_macro": 0.5947414485755244, | |
| "eval_f1_micro": 0.8362663373469179, | |
| "eval_loss": 0.30450889468193054, | |
| "eval_roc_auc": 0.8639863234543288, | |
| "eval_runtime": 72.0072, | |
| "eval_samples_per_second": 53.384, | |
| "eval_steps_per_second": 0.847, | |
| "learning_rate": 0.001, | |
| "step": 6516 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.27601456815816855, | |
| "eval_f1_macro": 0.599485078190572, | |
| "eval_f1_micro": 0.8396123222507695, | |
| "eval_loss": 0.30367332696914673, | |
| "eval_roc_auc": 0.8671612725761473, | |
| "eval_runtime": 69.1194, | |
| "eval_samples_per_second": 55.614, | |
| "eval_steps_per_second": 0.883, | |
| "learning_rate": 0.001, | |
| "step": 6697 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.2736732570239334, | |
| "eval_f1_macro": 0.5859770918782524, | |
| "eval_f1_micro": 0.838785705136416, | |
| "eval_loss": 0.3015189468860626, | |
| "eval_roc_auc": 0.8662036068197154, | |
| "eval_runtime": 67.6585, | |
| "eval_samples_per_second": 56.815, | |
| "eval_steps_per_second": 0.902, | |
| "learning_rate": 0.0001, | |
| "step": 6878 | |
| }, | |
| { | |
| "epoch": 38.67403314917127, | |
| "grad_norm": 0.17121317982673645, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3203, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.2736732570239334, | |
| "eval_f1_macro": 0.5995341133824836, | |
| "eval_f1_micro": 0.838145545925432, | |
| "eval_loss": 0.30049240589141846, | |
| "eval_roc_auc": 0.8656439298623401, | |
| "eval_runtime": 68.079, | |
| "eval_samples_per_second": 56.464, | |
| "eval_steps_per_second": 0.896, | |
| "learning_rate": 0.0001, | |
| "step": 7059 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.2695109261186264, | |
| "eval_f1_macro": 0.6125564265199016, | |
| "eval_f1_micro": 0.841743177276981, | |
| "eval_loss": 0.30099743604660034, | |
| "eval_roc_auc": 0.8691556197305788, | |
| "eval_runtime": 66.33, | |
| "eval_samples_per_second": 57.953, | |
| "eval_steps_per_second": 0.92, | |
| "learning_rate": 0.0001, | |
| "step": 7240 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.27419354838709675, | |
| "eval_f1_macro": 0.6073049013268815, | |
| "eval_f1_micro": 0.8403053435114504, | |
| "eval_loss": 0.29900264739990234, | |
| "eval_roc_auc": 0.8677200013927284, | |
| "eval_runtime": 70.4045, | |
| "eval_samples_per_second": 54.599, | |
| "eval_steps_per_second": 0.866, | |
| "learning_rate": 0.0001, | |
| "step": 7421 | |
| }, | |
| { | |
| "epoch": 41.43646408839779, | |
| "grad_norm": 0.16666004061698914, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3165, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.2713319458896982, | |
| "eval_f1_macro": 0.5991905306617578, | |
| "eval_f1_micro": 0.8408500229322733, | |
| "eval_loss": 0.2996482849121094, | |
| "eval_roc_auc": 0.8681113305915592, | |
| "eval_runtime": 64.5424, | |
| "eval_samples_per_second": 59.558, | |
| "eval_steps_per_second": 0.945, | |
| "learning_rate": 0.0001, | |
| "step": 7602 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.2695109261186264, | |
| "eval_f1_macro": 0.6091761352198065, | |
| "eval_f1_micro": 0.8414383822001469, | |
| "eval_loss": 0.29860275983810425, | |
| "eval_roc_auc": 0.8688227620059982, | |
| "eval_runtime": 62.4452, | |
| "eval_samples_per_second": 61.558, | |
| "eval_steps_per_second": 0.977, | |
| "learning_rate": 0.0001, | |
| "step": 7783 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.2749739854318418, | |
| "eval_f1_macro": 0.595393170760367, | |
| "eval_f1_micro": 0.83963159650068, | |
| "eval_loss": 0.2981945872306824, | |
| "eval_roc_auc": 0.8668373099533652, | |
| "eval_runtime": 63.2599, | |
| "eval_samples_per_second": 60.765, | |
| "eval_steps_per_second": 0.964, | |
| "learning_rate": 0.0001, | |
| "step": 7964 | |
| }, | |
| { | |
| "epoch": 44.19889502762431, | |
| "grad_norm": 0.15173059701919556, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3138, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.2757544224765869, | |
| "eval_f1_macro": 0.6028068266895004, | |
| "eval_f1_micro": 0.8400674260611943, | |
| "eval_loss": 0.2977070212364197, | |
| "eval_roc_auc": 0.8673735530917092, | |
| "eval_runtime": 63.0465, | |
| "eval_samples_per_second": 60.971, | |
| "eval_steps_per_second": 0.968, | |
| "learning_rate": 0.0001, | |
| "step": 8145 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.2754942767950052, | |
| "eval_f1_macro": 0.5966152252417688, | |
| "eval_f1_micro": 0.8406257200645179, | |
| "eval_loss": 0.29817572236061096, | |
| "eval_roc_auc": 0.8677318087609064, | |
| "eval_runtime": 63.3542, | |
| "eval_samples_per_second": 60.675, | |
| "eval_steps_per_second": 0.963, | |
| "learning_rate": 0.0001, | |
| "step": 8326 | |
| }, | |
| { | |
| "epoch": 46.96132596685083, | |
| "grad_norm": 0.19454629719257355, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3125, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.2767950052029136, | |
| "eval_f1_macro": 0.5893483505785619, | |
| "eval_f1_micro": 0.8377610088898078, | |
| "eval_loss": 0.29967373609542847, | |
| "eval_roc_auc": 0.8650328028655897, | |
| "eval_runtime": 66.3004, | |
| "eval_samples_per_second": 57.979, | |
| "eval_steps_per_second": 0.92, | |
| "learning_rate": 0.0001, | |
| "step": 8507 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.27471383975026015, | |
| "eval_f1_macro": 0.6135295162351735, | |
| "eval_f1_micro": 0.8420492820421911, | |
| "eval_loss": 0.2978098392486572, | |
| "eval_roc_auc": 0.8693551899664994, | |
| "eval_runtime": 64.6705, | |
| "eval_samples_per_second": 59.44, | |
| "eval_steps_per_second": 0.943, | |
| "learning_rate": 0.0001, | |
| "step": 8688 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.27471383975026015, | |
| "eval_f1_macro": 0.6017406274418097, | |
| "eval_f1_micro": 0.8398618395804016, | |
| "eval_loss": 0.29812732338905334, | |
| "eval_roc_auc": 0.8671341366526011, | |
| "eval_runtime": 62.9848, | |
| "eval_samples_per_second": 61.031, | |
| "eval_steps_per_second": 0.968, | |
| "learning_rate": 0.0001, | |
| "step": 8869 | |
| }, | |
| { | |
| "epoch": 49.72375690607735, | |
| "grad_norm": 0.1885526031255722, | |
| "learning_rate": 0.0001, | |
| "loss": 0.312, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.2702913631633715, | |
| "eval_f1_macro": 0.6112978412836683, | |
| "eval_f1_micro": 0.8410134158403285, | |
| "eval_loss": 0.29766079783439636, | |
| "eval_roc_auc": 0.8684402374503033, | |
| "eval_runtime": 65.3588, | |
| "eval_samples_per_second": 58.814, | |
| "eval_steps_per_second": 0.933, | |
| "learning_rate": 0.0001, | |
| "step": 9050 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.2715920915712799, | |
| "eval_f1_macro": 0.6110237777757073, | |
| "eval_f1_micro": 0.8419122264928279, | |
| "eval_loss": 0.30011385679244995, | |
| "eval_roc_auc": 0.8695847405483492, | |
| "eval_runtime": 69.6934, | |
| "eval_samples_per_second": 55.156, | |
| "eval_steps_per_second": 0.875, | |
| "learning_rate": 0.0001, | |
| "step": 9231 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.27887617065556713, | |
| "eval_f1_macro": 0.6011341201375136, | |
| "eval_f1_micro": 0.8379989163807117, | |
| "eval_loss": 0.29774174094200134, | |
| "eval_roc_auc": 0.8652877966213873, | |
| "eval_runtime": 70.0846, | |
| "eval_samples_per_second": 54.848, | |
| "eval_steps_per_second": 0.87, | |
| "learning_rate": 0.0001, | |
| "step": 9412 | |
| }, | |
| { | |
| "epoch": 52.48618784530387, | |
| "grad_norm": 0.20562438666820526, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3115, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.27289281997918835, | |
| "eval_f1_macro": 0.6151233978242734, | |
| "eval_f1_micro": 0.8425461059111437, | |
| "eval_loss": 0.2966245412826538, | |
| "eval_roc_auc": 0.8699284348856478, | |
| "eval_runtime": 67.2883, | |
| "eval_samples_per_second": 57.127, | |
| "eval_steps_per_second": 0.907, | |
| "learning_rate": 0.0001, | |
| "step": 9593 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.2809573361082206, | |
| "eval_f1_macro": 0.5974038753077745, | |
| "eval_f1_micro": 0.839878532091204, | |
| "eval_loss": 0.2976503372192383, | |
| "eval_roc_auc": 0.8669142710106849, | |
| "eval_runtime": 68.333, | |
| "eval_samples_per_second": 56.254, | |
| "eval_steps_per_second": 0.893, | |
| "learning_rate": 0.0001, | |
| "step": 9774 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.27237252861602496, | |
| "eval_f1_macro": 0.6118791964065535, | |
| "eval_f1_micro": 0.8407133757961783, | |
| "eval_loss": 0.2964698374271393, | |
| "eval_roc_auc": 0.8680032827448569, | |
| "eval_runtime": 66.9844, | |
| "eval_samples_per_second": 57.387, | |
| "eval_steps_per_second": 0.911, | |
| "learning_rate": 0.0001, | |
| "step": 9955 | |
| }, | |
| { | |
| "epoch": 55.248618784530386, | |
| "grad_norm": 0.16818420588970184, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3105, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.2786160249739854, | |
| "eval_f1_macro": 0.6058471375996606, | |
| "eval_f1_micro": 0.8408194622279129, | |
| "eval_loss": 0.29658928513526917, | |
| "eval_roc_auc": 0.8678807414102527, | |
| "eval_runtime": 69.2914, | |
| "eval_samples_per_second": 55.476, | |
| "eval_steps_per_second": 0.88, | |
| "learning_rate": 0.0001, | |
| "step": 10136 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.27471383975026015, | |
| "eval_f1_macro": 0.606792526005266, | |
| "eval_f1_micro": 0.8398854864270742, | |
| "eval_loss": 0.29776841402053833, | |
| "eval_roc_auc": 0.8671915469127291, | |
| "eval_runtime": 71.354, | |
| "eval_samples_per_second": 53.872, | |
| "eval_steps_per_second": 0.855, | |
| "learning_rate": 0.0001, | |
| "step": 10317 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.2721123829344433, | |
| "eval_f1_macro": 0.614564922504406, | |
| "eval_f1_micro": 0.8427313221299859, | |
| "eval_loss": 0.29647430777549744, | |
| "eval_roc_auc": 0.8699460215572541, | |
| "eval_runtime": 68.4, | |
| "eval_samples_per_second": 56.199, | |
| "eval_steps_per_second": 0.892, | |
| "learning_rate": 0.0001, | |
| "step": 10498 | |
| }, | |
| { | |
| "epoch": 58.011049723756905, | |
| "grad_norm": 0.2239256352186203, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3105, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.2793964620187305, | |
| "eval_f1_macro": 0.6005863231905055, | |
| "eval_f1_micro": 0.8405752518649542, | |
| "eval_loss": 0.2961284816265106, | |
| "eval_roc_auc": 0.8676383900900995, | |
| "eval_runtime": 74.2032, | |
| "eval_samples_per_second": 51.804, | |
| "eval_steps_per_second": 0.822, | |
| "learning_rate": 0.0001, | |
| "step": 10679 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.2796566077003122, | |
| "eval_f1_macro": 0.6113253981378993, | |
| "eval_f1_micro": 0.8428984624752626, | |
| "eval_loss": 0.29608383774757385, | |
| "eval_roc_auc": 0.8699772214199647, | |
| "eval_runtime": 67.9781, | |
| "eval_samples_per_second": 56.548, | |
| "eval_steps_per_second": 0.897, | |
| "learning_rate": 0.0001, | |
| "step": 10860 | |
| }, | |
| { | |
| "epoch": 60.773480662983424, | |
| "grad_norm": 0.20308265089988708, | |
| "learning_rate": 0.0001, | |
| "loss": 0.308, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.2796566077003122, | |
| "eval_f1_macro": 0.5998781229497869, | |
| "eval_f1_micro": 0.8415255756768527, | |
| "eval_loss": 0.2963137924671173, | |
| "eval_roc_auc": 0.8684482296406385, | |
| "eval_runtime": 65.3138, | |
| "eval_samples_per_second": 58.854, | |
| "eval_steps_per_second": 0.934, | |
| "learning_rate": 0.0001, | |
| "step": 11041 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.27887617065556713, | |
| "eval_f1_macro": 0.6017143414363826, | |
| "eval_f1_micro": 0.8405433111225014, | |
| "eval_loss": 0.29606395959854126, | |
| "eval_roc_auc": 0.8676237835288696, | |
| "eval_runtime": 66.6671, | |
| "eval_samples_per_second": 57.66, | |
| "eval_steps_per_second": 0.915, | |
| "learning_rate": 0.0001, | |
| "step": 11222 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.2736732570239334, | |
| "eval_f1_macro": 0.6107600812503934, | |
| "eval_f1_micro": 0.8421159560149325, | |
| "eval_loss": 0.29548379778862, | |
| "eval_roc_auc": 0.8692919777882774, | |
| "eval_runtime": 65.9628, | |
| "eval_samples_per_second": 58.275, | |
| "eval_steps_per_second": 0.925, | |
| "learning_rate": 0.0001, | |
| "step": 11403 | |
| }, | |
| { | |
| "epoch": 63.53591160220994, | |
| "grad_norm": 0.22083976864814758, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3083, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.2791363163371488, | |
| "eval_f1_macro": 0.6127112192164306, | |
| "eval_f1_micro": 0.8407360081778686, | |
| "eval_loss": 0.2951599955558777, | |
| "eval_roc_auc": 0.8678962734976112, | |
| "eval_runtime": 67.2265, | |
| "eval_samples_per_second": 57.18, | |
| "eval_steps_per_second": 0.907, | |
| "learning_rate": 0.0001, | |
| "step": 11584 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.27809573361082207, | |
| "eval_f1_macro": 0.6022403129456154, | |
| "eval_f1_micro": 0.839130323176836, | |
| "eval_loss": 0.29839980602264404, | |
| "eval_roc_auc": 0.8664061119366814, | |
| "eval_runtime": 69.4982, | |
| "eval_samples_per_second": 55.311, | |
| "eval_steps_per_second": 0.878, | |
| "learning_rate": 0.0001, | |
| "step": 11765 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.2739334027055151, | |
| "eval_f1_macro": 0.6104238923347367, | |
| "eval_f1_micro": 0.841473646741756, | |
| "eval_loss": 0.2956618666648865, | |
| "eval_roc_auc": 0.8687138112478329, | |
| "eval_runtime": 65.4466, | |
| "eval_samples_per_second": 58.735, | |
| "eval_steps_per_second": 0.932, | |
| "learning_rate": 0.0001, | |
| "step": 11946 | |
| }, | |
| { | |
| "epoch": 66.29834254143647, | |
| "grad_norm": 0.245449036359787, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3051, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.27601456815816855, | |
| "eval_f1_macro": 0.6142232564356239, | |
| "eval_f1_micro": 0.841559361481444, | |
| "eval_loss": 0.29616212844848633, | |
| "eval_roc_auc": 0.868995015401048, | |
| "eval_runtime": 63.2029, | |
| "eval_samples_per_second": 60.82, | |
| "eval_steps_per_second": 0.965, | |
| "learning_rate": 0.0001, | |
| "step": 12127 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.277315296566077, | |
| "eval_f1_macro": 0.6083667891934388, | |
| "eval_f1_micro": 0.8412997050747482, | |
| "eval_loss": 0.29670244455337524, | |
| "eval_roc_auc": 0.8685716965013714, | |
| "eval_runtime": 62.807, | |
| "eval_samples_per_second": 61.203, | |
| "eval_steps_per_second": 0.971, | |
| "learning_rate": 0.0001, | |
| "step": 12308 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.27289281997918835, | |
| "eval_f1_macro": 0.6160532530170337, | |
| "eval_f1_micro": 0.8406188667703209, | |
| "eval_loss": 0.29603201150894165, | |
| "eval_roc_auc": 0.8679078999484645, | |
| "eval_runtime": 66.3361, | |
| "eval_samples_per_second": 57.947, | |
| "eval_steps_per_second": 0.92, | |
| "learning_rate": 0.0001, | |
| "step": 12489 | |
| }, | |
| { | |
| "epoch": 69.06077348066299, | |
| "grad_norm": 0.240888312458992, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3066, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.268210197710718, | |
| "eval_f1_macro": 0.6247633367937411, | |
| "eval_f1_micro": 0.8433517310587054, | |
| "eval_loss": 0.2971898317337036, | |
| "eval_roc_auc": 0.8708423261877216, | |
| "eval_runtime": 65.1371, | |
| "eval_samples_per_second": 59.014, | |
| "eval_steps_per_second": 0.936, | |
| "learning_rate": 0.0001, | |
| "step": 12670 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.28017689906347554, | |
| "eval_f1_macro": 0.6081264630991828, | |
| "eval_f1_micro": 0.8396197327852004, | |
| "eval_loss": 0.2964514493942261, | |
| "eval_roc_auc": 0.8667672572575927, | |
| "eval_runtime": 66.6789, | |
| "eval_samples_per_second": 57.649, | |
| "eval_steps_per_second": 0.915, | |
| "learning_rate": 1e-05, | |
| "step": 12851 | |
| }, | |
| { | |
| "epoch": 71.8232044198895, | |
| "grad_norm": 0.2170819193124771, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3061, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.28069719042663893, | |
| "eval_f1_macro": 0.603429563858726, | |
| "eval_f1_micro": 0.8422399017676703, | |
| "eval_loss": 0.29609331488609314, | |
| "eval_roc_auc": 0.8690857705667967, | |
| "eval_runtime": 67.2956, | |
| "eval_samples_per_second": 57.121, | |
| "eval_steps_per_second": 0.906, | |
| "learning_rate": 1e-05, | |
| "step": 13032 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.2809573361082206, | |
| "eval_f1_macro": 0.6080269607484758, | |
| "eval_f1_micro": 0.8408677611863842, | |
| "eval_loss": 0.2954128384590149, | |
| "eval_roc_auc": 0.867912866799092, | |
| "eval_runtime": 67.8811, | |
| "eval_samples_per_second": 56.628, | |
| "eval_steps_per_second": 0.899, | |
| "learning_rate": 1e-05, | |
| "step": 13213 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.2809573361082206, | |
| "eval_f1_macro": 0.6093453328873296, | |
| "eval_f1_micro": 0.8411459000025601, | |
| "eval_loss": 0.29537200927734375, | |
| "eval_roc_auc": 0.8681591237557506, | |
| "eval_runtime": 65.6885, | |
| "eval_samples_per_second": 58.519, | |
| "eval_steps_per_second": 0.929, | |
| "learning_rate": 1e-05, | |
| "step": 13394 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "learning_rate": 1e-05, | |
| "step": 13394, | |
| "total_flos": 1.9211796047151268e+19, | |
| "train_loss": 0.3220266872187844, | |
| "train_runtime": 23053.7141, | |
| "train_samples_per_second": 49.979, | |
| "train_steps_per_second": 0.785 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 18100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 10, | |
| "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": 1.9211796047151268e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |