End of training

README.md

@@ -0,0 +1,116 @@
+---
+base_model: napsternxg/gte-small-L3-ingredient-v2
+tags:
+- generated_from_trainer
+datasets:
+- nyt_ingredients
+model-index:
+- name: nyt_ingredients-tagger-gte-small-L3-ingredient-v2
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# nyt_ingredients-tagger-gte-small-L3-ingredient-v2
+
+This model is a fine-tuned version of [napsternxg/gte-small-L3-ingredient-v2](https://huggingface.co/napsternxg/gte-small-L3-ingredient-v2) on the nyt_ingredients dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4705
+- Comment: {'precision': 0.6653720111664037, 'recall': 0.7644679960953842, 'f1': 0.7114860480207656, 'number': 7171}
+- Name: {'precision': 0.7963734889537307, 'recall': 0.8212788823213326, 'f1': 0.8086344637849849, 'number': 9305}
+- Qty: {'precision': 0.981755684822845, 'recall': 0.9907938625750501, 'f1': 0.9862540673351484, 'number': 7495}
+- Range End: {'precision': 0.6240601503759399, 'recall': 0.9222222222222223, 'f1': 0.7443946188340806, 'number': 90}
+- Unit: {'precision': 0.923100969052829, 'recall': 0.9861412589747871, 'f1': 0.95358036651328, 'number': 5989}
+- Overall Precision: 0.8312
+- Overall Recall: 0.8832
+- Overall F1: 0.8564
+- Overall Accuracy: 0.8350
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 10
+
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Comment                                                                                                   | Name                                                                                                      | Qty                                                                                                       | Range End                                                                                               | Unit                                                                                                      | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:---------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------:|:-------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
+| 0.6931        | 0.2   | 1000  | 0.6230          | {'precision': 0.5236127508854782, 'recall': 0.6426604839878278, 'f1': 0.5770606987183658, 'number': 6901} | {'precision': 0.7773012207192346, 'recall': 0.7980128711753416, 'f1': 0.7875208913649026, 'number': 8857} | {'precision': 0.9824561403508771, 'recall': 0.9737098344693281, 'f1': 0.9780634343998882, 'number': 7189} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 92}                                              | {'precision': 0.9192320423700762, 'recall': 0.9689462665736218, 'f1': 0.9434346865975879, 'number': 5732} | 0.7829            | 0.8362         | 0.8086     | 0.7986           |
+| 0.5826        | 0.4   | 2000  | 0.5531          | {'precision': 0.5435365427205968, 'recall': 0.6756991740327489, 'f1': 0.602454780361757, 'number': 6901}  | {'precision': 0.7757141305528402, 'recall': 0.8063678446426555, 'f1': 0.7907440212577502, 'number': 8857} | {'precision': 0.9659028914348063, 'recall': 0.9851161496731117, 'f1': 0.9754149163280765, 'number': 7189} | {'precision': 0.6055045871559633, 'recall': 0.717391304347826, 'f1': 0.6567164179104478, 'number': 92}  | {'precision': 0.9177049180327869, 'recall': 0.9766224703419399, 'f1': 0.9462474645030426, 'number': 5732} | 0.7837            | 0.8533         | 0.8170     | 0.8079           |
+| 0.5435        | 0.59  | 3000  | 0.5327          | {'precision': 0.5801165331391115, 'recall': 0.6925083321257789, 'f1': 0.6313494946826079, 'number': 6901} | {'precision': 0.7839085513299626, 'recall': 0.8052387941740996, 'f1': 0.7944305207463102, 'number': 8857} | {'precision': 0.9771436487048067, 'recall': 0.9812213103352344, 'f1': 0.9791782343142699, 'number': 7189} | {'precision': 0.6302521008403361, 'recall': 0.8152173913043478, 'f1': 0.7109004739336493, 'number': 92} | {'precision': 0.9126637554585153, 'recall': 0.9844731332868109, 'f1': 0.9472093999160722, 'number': 5732} | 0.7999            | 0.8579         | 0.8279     | 0.8150           |
+| 0.5333        | 0.79  | 4000  | 0.5212          | {'precision': 0.5875760104924288, 'recall': 0.7140994058832053, 'f1': 0.6446886446886447, 'number': 6901} | {'precision': 0.786660777385159, 'recall': 0.8043355537992548, 'f1': 0.7953999888349244, 'number': 8857}  | {'precision': 0.9837716843872412, 'recall': 0.9781610794269022, 'f1': 0.9809583594894328, 'number': 7189} | {'precision': 0.6390977443609023, 'recall': 0.9239130434782609, 'f1': 0.7555555555555556, 'number': 92} | {'precision': 0.9263366992219831, 'recall': 0.9762735519888346, 'f1': 0.9506497918967128, 'number': 5732} | 0.8050            | 0.8608         | 0.8319     | 0.8183           |
+| 0.5199        | 0.99  | 5000  | 0.5121          | {'precision': 0.5873457536898137, 'recall': 0.7035212288074192, 'f1': 0.6402057097646205, 'number': 6901} | {'precision': 0.7890984865101996, 'recall': 0.812351812126002, 'f1': 0.8005563282336579, 'number': 8857}  | {'precision': 0.978519955654102, 'recall': 0.9821950201697037, 'f1': 0.9803540437348144, 'number': 7189}  | {'precision': 0.6615384615384615, 'recall': 0.9347826086956522, 'f1': 0.7747747747747747, 'number': 92} | {'precision': 0.9197843489625878, 'recall': 0.9822051639916259, 'f1': 0.9499704716105627, 'number': 5732} | 0.8047            | 0.8629         | 0.8328     | 0.8186           |
+| 0.4983        | 1.19  | 6000  | 0.5088          | {'precision': 0.6041901894367081, 'recall': 0.6978698739313143, 'f1': 0.6476600322754169, 'number': 6901} | {'precision': 0.7801047120418848, 'recall': 0.8074968951112115, 'f1': 0.7935644937586686, 'number': 8857} | {'precision': 0.980998613037448, 'recall': 0.983864237028794, 'f1': 0.9824293353705119, 'number': 7189}   | {'precision': 0.6511627906976745, 'recall': 0.9130434782608695, 'f1': 0.7601809954751132, 'number': 92} | {'precision': 0.9193864229765013, 'recall': 0.9829030006978368, 'f1': 0.9500843170320405, 'number': 5732} | 0.8090            | 0.8606         | 0.8340     | 0.8192           |
+| 0.5031        | 1.39  | 7000  | 0.5031          | {'precision': 0.6055345911949686, 'recall': 0.6975800608607449, 'f1': 0.6483065113460373, 'number': 6901} | {'precision': 0.7835422343324251, 'recall': 0.8116743818448685, 'f1': 0.797360248447205, 'number': 8857}  | {'precision': 0.9752815160670145, 'recall': 0.9878981777715955, 'f1': 0.9815493055075669, 'number': 7189} | {'precision': 0.6666666666666666, 'recall': 0.9130434782608695, 'f1': 0.7706422018348623, 'number': 92} | {'precision': 0.9269467043001827, 'recall': 0.9740055826936497, 'f1': 0.9498936622713738, 'number': 5732} | 0.8107            | 0.8610         | 0.8351     | 0.8199           |
+| 0.492         | 1.58  | 8000  | 0.5041          | {'precision': 0.6209482341557814, 'recall': 0.743950152151862, 'f1': 0.6769068494956819, 'number': 6901}  | {'precision': 0.7778260869565218, 'recall': 0.8079485152986339, 'f1': 0.7926012072880323, 'number': 8857} | {'precision': 0.983001254005852, 'recall': 0.9813604117401585, 'f1': 0.9821801475706529, 'number': 7189}  | {'precision': 0.6423357664233577, 'recall': 0.9565217391304348, 'f1': 0.7685589519650656, 'number': 92} | {'precision': 0.9215493188905302, 'recall': 0.9795882763433357, 'f1': 0.9496828752642706, 'number': 5732} | 0.8113            | 0.8706         | 0.8399     | 0.8246           |
+| 0.4817        | 1.78  | 9000  | 0.4901          | {'precision': 0.6196705426356589, 'recall': 0.7413418345167367, 'f1': 0.6750676255195619, 'number': 6901} | {'precision': 0.7891447368421053, 'recall': 0.8125776222197132, 'f1': 0.8006897702620014, 'number': 8857} | {'precision': 0.9798370390829996, 'recall': 0.9869244679371262, 'f1': 0.9833679833679833, 'number': 7189} | {'precision': 0.6474820143884892, 'recall': 0.9782608695652174, 'f1': 0.7792207792207793, 'number': 92} | {'precision': 0.9168422758956071, 'recall': 0.9867411025819958, 'f1': 0.9505083606419629, 'number': 5732} | 0.8134            | 0.8743         | 0.8427     | 0.8293           |
+| 0.4905        | 1.98  | 10000 | 0.4892          | {'precision': 0.6241463414634146, 'recall': 0.7416316475873062, 'f1': 0.6778359049069598, 'number': 6901} | {'precision': 0.7911287653095002, 'recall': 0.8095291859546122, 'f1': 0.8002232142857143, 'number': 8857} | {'precision': 0.9846860643185299, 'recall': 0.983864237028794, 'f1': 0.9842749791260784, 'number': 7189}  | {'precision': 0.6717557251908397, 'recall': 0.9565217391304348, 'f1': 0.789237668161435, 'number': 92}  | {'precision': 0.9170723790976956, 'recall': 0.9858688066992324, 'f1': 0.9502270052127123, 'number': 5732} | 0.8166            | 0.8724         | 0.8436     | 0.8288           |
+| 0.4757        | 2.18  | 11000 | 0.4954          | {'precision': 0.6271186440677966, 'recall': 0.7398927691638892, 'f1': 0.6788539520042544, 'number': 6901} | {'precision': 0.7921020789792103, 'recall': 0.8130292424071356, 'f1': 0.802429240026744, 'number': 8857}  | {'precision': 0.9801077496891836, 'recall': 0.9869244679371262, 'f1': 0.9835042971998891, 'number': 7189} | {'precision': 0.6428571428571429, 'recall': 0.9782608695652174, 'f1': 0.7758620689655172, 'number': 92} | {'precision': 0.9226609864001311, 'recall': 0.9823796231681786, 'f1': 0.9515842839036756, 'number': 5732} | 0.8179            | 0.8732         | 0.8447     | 0.8274           |
+| 0.4691        | 2.38  | 12000 | 0.4903          | {'precision': 0.6292872780345414, 'recall': 0.7497464135632517, 'f1': 0.6842557693579316, 'number': 6901} | {'precision': 0.7935547734271887, 'recall': 0.8146099130631139, 'f1': 0.803944509443423, 'number': 8857}  | {'precision': 0.9806816613771215, 'recall': 0.9885936847962165, 'f1': 0.984621778886118, 'number': 7189}  | {'precision': 0.6544117647058824, 'recall': 0.967391304347826, 'f1': 0.780701754385965, 'number': 92}   | {'precision': 0.9238204456094364, 'recall': 0.9837752965806001, 'f1': 0.9528556944913822, 'number': 5732} | 0.8189            | 0.8767         | 0.8468     | 0.8301           |
+| 0.4765        | 2.57  | 13000 | 0.4846          | {'precision': 0.6338439095550693, 'recall': 0.7555426749746413, 'f1': 0.6893633899649634, 'number': 6901} | {'precision': 0.7977640026566305, 'recall': 0.8137066726882691, 'f1': 0.805656475322788, 'number': 8857}  | {'precision': 0.9850891861761427, 'recall': 0.9833078314090973, 'f1': 0.9841977027497391, 'number': 7189} | {'precision': 0.6774193548387096, 'recall': 0.9130434782608695, 'f1': 0.7777777777777777, 'number': 92} | {'precision': 0.9219950940310712, 'recall': 0.9836008374040475, 'f1': 0.9518021440027011, 'number': 5732} | 0.8219            | 0.8763         | 0.8482     | 0.8307           |
+| 0.4747        | 2.77  | 14000 | 0.4844          | {'precision': 0.6380844929171368, 'recall': 0.7375742645993334, 'f1': 0.6842317515795134, 'number': 6901} | {'precision': 0.791246160596753, 'recall': 0.8143841029694028, 'f1': 0.8026484170700496, 'number': 8857}  | {'precision': 0.9796255506607929, 'recall': 0.9898455974405341, 'f1': 0.9847090569431951, 'number': 7189} | {'precision': 0.6546762589928058, 'recall': 0.9891304347826086, 'f1': 0.7878787878787878, 'number': 92} | {'precision': 0.9195252804422045, 'recall': 0.9867411025819958, 'f1': 0.9519481612387444, 'number': 5732} | 0.8212            | 0.8747         | 0.8471     | 0.8314           |
+| 0.4654        | 2.97  | 15000 | 0.4808          | {'precision': 0.6453574975173784, 'recall': 0.7533690769453703, 'f1': 0.6951928862739855, 'number': 6901} | {'precision': 0.7925608011444921, 'recall': 0.8131421474539912, 'f1': 0.8027195720017833, 'number': 8857} | {'precision': 0.9827013562136728, 'recall': 0.9877590763666713, 'f1': 0.9852237252861602, 'number': 7189} | {'precision': 0.6641221374045801, 'recall': 0.9456521739130435, 'f1': 0.7802690582959643, 'number': 92} | {'precision': 0.9255791030064071, 'recall': 0.9829030006978368, 'f1': 0.9533801506049582, 'number': 5732} | 0.8246            | 0.8767         | 0.8499     | 0.8332           |
+| 0.4586        | 3.17  | 16000 | 0.4827          | {'precision': 0.6517362858580775, 'recall': 0.7506158527749601, 'f1': 0.6976900801400768, 'number': 6901} | {'precision': 0.7901383703052932, 'recall': 0.812351812126002, 'f1': 0.8010911317708622, 'number': 8857}  | {'precision': 0.979906413432425, 'recall': 0.990402003060231, 'f1': 0.9851262538913871, 'number': 7189}   | {'precision': 0.6641221374045801, 'recall': 0.9456521739130435, 'f1': 0.7802690582959643, 'number': 92} | {'precision': 0.9193364774760123, 'recall': 0.9862177250523377, 'f1': 0.9516034003871727, 'number': 5732} | 0.8247            | 0.8771         | 0.8501     | 0.8317           |
+| 0.4612        | 3.37  | 17000 | 0.4869          | {'precision': 0.6441053408811224, 'recall': 0.7584408056803362, 'f1': 0.6966127636920211, 'number': 6901} | {'precision': 0.7845718637797846, 'recall': 0.8141582928756915, 'f1': 0.7990913120567377, 'number': 8857} | {'precision': 0.9844271412680756, 'recall': 0.9848379468632633, 'f1': 0.9846325012168833, 'number': 7189} | {'precision': 0.6641221374045801, 'recall': 0.9456521739130435, 'f1': 0.7802690582959643, 'number': 92} | {'precision': 0.9158439877002751, 'recall': 0.9872644801116539, 'f1': 0.9502140878179833, 'number': 5732} | 0.8200            | 0.8783         | 0.8482     | 0.8308           |
+| 0.4546        | 3.56  | 18000 | 0.4806          | {'precision': 0.6482188295165394, 'recall': 0.7382987972757571, 'f1': 0.6903326332904275, 'number': 6901} | {'precision': 0.788654181184669, 'recall': 0.8177712543750706, 'f1': 0.8029488387561666, 'number': 8857}  | {'precision': 0.9817704736914791, 'recall': 0.9888718876060648, 'f1': 0.9853083853083853, 'number': 7189} | {'precision': 0.6854838709677419, 'recall': 0.9239130434782609, 'f1': 0.787037037037037, 'number': 92}  | {'precision': 0.9236179260713118, 'recall': 0.9851709699930217, 'f1': 0.9534019922336655, 'number': 5732} | 0.8249            | 0.8752         | 0.8493     | 0.8315           |
+| 0.4471        | 3.76  | 19000 | 0.4823          | {'precision': 0.6466100652147164, 'recall': 0.7614838429213158, 'f1': 0.6993611924407772, 'number': 6901} | {'precision': 0.7929254092057564, 'recall': 0.8149486282036807, 'f1': 0.8037861915367484, 'number': 8857} | {'precision': 0.983906770255272, 'recall': 0.9865071637223536, 'f1': 0.9852052510939778, 'number': 7189}  | {'precision': 0.676923076923077, 'recall': 0.9565217391304348, 'f1': 0.7927927927927929, 'number': 92}  | {'precision': 0.9211943220753793, 'recall': 0.984996510816469, 'f1': 0.9520276536548351, 'number': 5732}  | 0.8242            | 0.8793         | 0.8508     | 0.8328           |
+| 0.4542        | 3.96  | 20000 | 0.4758          | {'precision': 0.6508980521123198, 'recall': 0.7456890305752789, 'f1': 0.6950766529344229, 'number': 6901} | {'precision': 0.7868081382629621, 'recall': 0.8121260020322908, 'f1': 0.7992666259236624, 'number': 8857} | {'precision': 0.9829734219269103, 'recall': 0.9877590763666713, 'f1': 0.9853604384930272, 'number': 7189} | {'precision': 0.6744186046511628, 'recall': 0.9456521739130435, 'f1': 0.7873303167420815, 'number': 92} | {'precision': 0.920559986977047, 'recall': 0.9865666434054431, 'f1': 0.9524210526315788, 'number': 5732}  | 0.8244            | 0.8753         | 0.8491     | 0.8326           |
+| 0.4452        | 4.16  | 21000 | 0.4833          | {'precision': 0.6566687938736439, 'recall': 0.7455441240399941, 'f1': 0.6982899022801302, 'number': 6901} | {'precision': 0.7843478260869565, 'recall': 0.8147228181099695, 'f1': 0.7992468294844105, 'number': 8857} | {'precision': 0.981750311074243, 'recall': 0.9877590763666713, 'f1': 0.9847455276660656, 'number': 7189}  | {'precision': 0.6854838709677419, 'recall': 0.9239130434782609, 'f1': 0.787037037037037, 'number': 92}  | {'precision': 0.9249303392886412, 'recall': 0.9844731332868109, 'f1': 0.9537733457280487, 'number': 5732} | 0.8261            | 0.8755         | 0.8501     | 0.8339           |
+| 0.4412        | 4.36  | 22000 | 0.4888          | {'precision': 0.6534954407294833, 'recall': 0.7477177220692653, 'f1': 0.6974386700006758, 'number': 6901} | {'precision': 0.7870067582297798, 'recall': 0.8151744382973919, 'f1': 0.8008429926238145, 'number': 8857} | {'precision': 0.9839268394069558, 'recall': 0.9877590763666713, 'f1': 0.9858392336526447, 'number': 7189} | {'precision': 0.6717557251908397, 'recall': 0.9565217391304348, 'f1': 0.789237668161435, 'number': 92}  | {'precision': 0.9248280379954144, 'recall': 0.9851709699930217, 'f1': 0.9540462916033113, 'number': 5732} | 0.8261            | 0.8764         | 0.8505     | 0.8325           |
+| 0.4423        | 4.55  | 23000 | 0.4827          | {'precision': 0.6532828282828282, 'recall': 0.7497464135632517, 'f1': 0.6981985021253626, 'number': 6901} | {'precision': 0.7874007180937874, 'recall': 0.817093824093937, 'f1': 0.8019725177304966, 'number': 8857}  | {'precision': 0.9835202880487467, 'recall': 0.9878981777715955, 'f1': 0.9857043719639139, 'number': 7189} | {'precision': 0.6766917293233082, 'recall': 0.9782608695652174, 'f1': 0.7999999999999999, 'number': 92} | {'precision': 0.9264850672792911, 'recall': 0.984996510816469, 'f1': 0.9548452562151193, 'number': 5732}  | 0.8262            | 0.8776         | 0.8511     | 0.8338           |
+| 0.4507        | 4.75  | 24000 | 0.4740          | {'precision': 0.6584520600693108, 'recall': 0.7433705260107231, 'f1': 0.69833923223523, 'number': 6901}   | {'precision': 0.7933530766699572, 'recall': 0.8166422039065147, 'f1': 0.8048291977300545, 'number': 8857} | {'precision': 0.9801898472967395, 'recall': 0.9910975100848518, 'f1': 0.9856135011758197, 'number': 7189} | {'precision': 0.6692307692307692, 'recall': 0.9456521739130435, 'f1': 0.7837837837837838, 'number': 92} | {'precision': 0.922976501305483, 'recall': 0.9867411025819958, 'f1': 0.9537942664418213, 'number': 5732}  | 0.8290            | 0.8770         | 0.8523     | 0.8347           |
+| 0.4452        | 4.95  | 25000 | 0.4791          | {'precision': 0.6528792569659443, 'recall': 0.7639472540211564, 'f1': 0.7040598290598291, 'number': 6901} | {'precision': 0.7875054513737462, 'recall': 0.8155131534379587, 'f1': 0.8012646292084975, 'number': 8857} | {'precision': 0.9825798423890502, 'recall': 0.9885936847962165, 'f1': 0.9855775897933713, 'number': 7189} | {'precision': 0.6640625, 'recall': 0.9239130434782609, 'f1': 0.7727272727272728, 'number': 92}          | {'precision': 0.9227881162259223, 'recall': 0.9862177250523377, 'f1': 0.953449148254343, 'number': 5732}  | 0.8245            | 0.8807         | 0.8517     | 0.8352           |
+| 0.4274        | 5.15  | 26000 | 0.4807          | {'precision': 0.6607946026986506, 'recall': 0.766410665120997, 'f1': 0.7096947333109694, 'number': 6901}  | {'precision': 0.795727342803656, 'recall': 0.8158518685785254, 'f1': 0.8056639536180176, 'number': 8857}  | {'precision': 0.9844660194174757, 'recall': 0.9873417721518988, 'f1': 0.9859017987360232, 'number': 7189} | {'precision': 0.6935483870967742, 'recall': 0.9347826086956522, 'f1': 0.7962962962962964, 'number': 92} | {'precision': 0.9229010127409344, 'recall': 0.9856943475226797, 'f1': 0.9532647207693606, 'number': 5732} | 0.8300            | 0.8811         | 0.8548     | 0.8361           |
+| 0.4312        | 5.35  | 27000 | 0.4861          | {'precision': 0.6542461005199307, 'recall': 0.7658310389798579, 'f1': 0.7056545830829829, 'number': 6901} | {'precision': 0.7903508771929825, 'recall': 0.8138195777351248, 'f1': 0.8019135562107137, 'number': 8857} | {'precision': 0.981501932633904, 'recall': 0.989010989010989, 'f1': 0.9852421533984619, 'number': 7189}   | {'precision': 0.6717557251908397, 'recall': 0.9565217391304348, 'f1': 0.789237668161435, 'number': 92}  | {'precision': 0.9154405820533549, 'recall': 0.9877878576413119, 'f1': 0.9502391541495343, 'number': 5732} | 0.8243            | 0.8812         | 0.8518     | 0.8335           |
+| 0.4315        | 5.54  | 28000 | 0.4820          | {'precision': 0.6583983883152859, 'recall': 0.7577162730039124, 'f1': 0.7045745469244761, 'number': 6901} | {'precision': 0.7930996593780903, 'recall': 0.8149486282036807, 'f1': 0.8038757099899767, 'number': 8857} | {'precision': 0.9831328632655882, 'recall': 0.9891500904159132, 'f1': 0.9861322978782415, 'number': 7189} | {'precision': 0.6796875, 'recall': 0.9456521739130435, 'f1': 0.7909090909090909, 'number': 92}          | {'precision': 0.9204878048780488, 'recall': 0.9876133984647593, 'f1': 0.9528698872243729, 'number': 5732} | 0.8282            | 0.8796         | 0.8531     | 0.8341           |
+| 0.4316        | 5.74  | 29000 | 0.4762          | {'precision': 0.6612373737373738, 'recall': 0.7588755252861904, 'f1': 0.7066999527697188, 'number': 6901} | {'precision': 0.7980441709702231, 'recall': 0.8200293553121825, 'f1': 0.8088874039425326, 'number': 8857} | {'precision': 0.9830040071852978, 'recall': 0.9895673946306858, 'f1': 0.9862747816442535, 'number': 7189} | {'precision': 0.672, 'recall': 0.9130434782608695, 'f1': 0.7741935483870968, 'number': 92}              | {'precision': 0.9263813739957371, 'recall': 0.9856943475226797, 'f1': 0.955117910573916, 'number': 5732}  | 0.8316            | 0.8810         | 0.8556     | 0.8370           |
+| 0.4278        | 5.94  | 30000 | 0.4823          | {'precision': 0.6596223583843942, 'recall': 0.7643819736270105, 'f1': 0.7081487447979596, 'number': 6901} | {'precision': 0.7918309242225142, 'recall': 0.8164163938128034, 'f1': 0.8039357385068653, 'number': 8857} | {'precision': 0.9835339698353397, 'recall': 0.9887327862011406, 'f1': 0.9861265260821309, 'number': 7189} | {'precision': 0.68, 'recall': 0.9239130434782609, 'f1': 0.783410138248848, 'number': 92}                | {'precision': 0.9244109947643979, 'recall': 0.9856943475226797, 'f1': 0.9540695710908478, 'number': 5732} | 0.8286            | 0.8811         | 0.8540     | 0.8348           |
+| 0.4211        | 6.14  | 31000 | 0.4794          | {'precision': 0.6637281206081512, 'recall': 0.7527894508042313, 'f1': 0.705458989679522, 'number': 6901}  | {'precision': 0.7826885671061162, 'recall': 0.8177712543750706, 'f1': 0.7998453978245265, 'number': 8857} | {'precision': 0.9804649883065071, 'recall': 0.9913757128947003, 'f1': 0.9858901646147462, 'number': 7189} | {'precision': 0.6854838709677419, 'recall': 0.9239130434782609, 'f1': 0.787037037037037, 'number': 92}  | {'precision': 0.925495333224169, 'recall': 0.9860432658757851, 'f1': 0.9548103724976772, 'number': 5732}  | 0.8274            | 0.8794         | 0.8526     | 0.8356           |
+| 0.4231        | 6.34  | 32000 | 0.4757          | {'precision': 0.6627510420613869, 'recall': 0.7603245906390378, 'f1': 0.7081927385612093, 'number': 6901} | {'precision': 0.7868816735672259, 'recall': 0.815400248391103, 'f1': 0.8008871638480732, 'number': 8857}  | {'precision': 0.9844875346260388, 'recall': 0.9887327862011406, 'f1': 0.9866055937261433, 'number': 7189} | {'precision': 0.696, 'recall': 0.9456521739130435, 'f1': 0.8018433179723502, 'number': 92}              | {'precision': 0.9235918367346939, 'recall': 0.9869155617585486, 'f1': 0.9542042675212956, 'number': 5732} | 0.8284            | 0.8801         | 0.8535     | 0.8364           |
+| 0.4302        | 6.53  | 33000 | 0.4786          | {'precision': 0.6624333248666497, 'recall': 0.7558324880452109, 'f1': 0.706057529610829, 'number': 6901}  | {'precision': 0.7953421948808085, 'recall': 0.8174325392345038, 'f1': 0.8062360801781736, 'number': 8857} | {'precision': 0.9832664914949523, 'recall': 0.989010989010989, 'f1': 0.986130374479889, 'number': 7189}   | {'precision': 0.7017543859649122, 'recall': 0.8695652173913043, 'f1': 0.7766990291262136, 'number': 92} | {'precision': 0.9243216737495914, 'recall': 0.9865666434054431, 'f1': 0.9544303797468354, 'number': 5732} | 0.8312            | 0.8794         | 0.8546     | 0.8355           |
+| 0.4223        | 6.73  | 34000 | 0.4795          | {'precision': 0.6597820577800304, 'recall': 0.7545283292276481, 'f1': 0.703981612925032, 'number': 6901}  | {'precision': 0.7922049485439019, 'recall': 0.8169809190470814, 'f1': 0.8044022011005503, 'number': 8857} | {'precision': 0.9832664914949523, 'recall': 0.989010989010989, 'f1': 0.986130374479889, 'number': 7189}   | {'precision': 0.6904761904761905, 'recall': 0.9456521739130435, 'f1': 0.7981651376146789, 'number': 92} | {'precision': 0.923730197615548, 'recall': 0.9867411025819958, 'f1': 0.9541965415436525, 'number': 5732}  | 0.8292            | 0.8792         | 0.8535     | 0.8363           |
+| 0.4184        | 6.93  | 35000 | 0.4803          | {'precision': 0.6614741641337386, 'recall': 0.7568468337922041, 'f1': 0.7059539095762654, 'number': 6901} | {'precision': 0.788057099269914, 'recall': 0.8165292988596591, 'f1': 0.8020405899966729, 'number': 8857}  | {'precision': 0.9823472624465591, 'recall': 0.9908193072750034, 'f1': 0.9865650969529086, 'number': 7189} | {'precision': 0.6821705426356589, 'recall': 0.9565217391304348, 'f1': 0.7963800904977375, 'number': 92} | {'precision': 0.9216229428059313, 'recall': 0.9867411025819958, 'f1': 0.9530710253601821, 'number': 5732} | 0.8278            | 0.8801         | 0.8532     | 0.8356           |
+| 0.4071        | 7.13  | 36000 | 0.4882          | {'precision': 0.6610340479192938, 'recall': 0.7596000579626141, 'f1': 0.7068977142471848, 'number': 6901} | {'precision': 0.7952072111685171, 'recall': 0.8167551089533702, 'f1': 0.8058371393561323, 'number': 8857} | {'precision': 0.9823423920540765, 'recall': 0.9905411044651551, 'f1': 0.9864247125640672, 'number': 7189} | {'precision': 0.6904761904761905, 'recall': 0.9456521739130435, 'f1': 0.7981651376146789, 'number': 92} | {'precision': 0.9252826478780928, 'recall': 0.9851709699930217, 'f1': 0.9542881284326151, 'number': 5732} | 0.8304            | 0.8804         | 0.8547     | 0.8358           |
+| 0.4093        | 7.33  | 37000 | 0.4856          | {'precision': 0.6579605098725319, 'recall': 0.7629329082741632, 'f1': 0.7065691471515801, 'number': 6901} | {'precision': 0.794249972566663, 'recall': 0.8172067291407926, 'f1': 0.8055648302726766, 'number': 8857}  | {'precision': 0.9832734310201825, 'recall': 0.9894282932257615, 'f1': 0.9863412604867227, 'number': 7189} | {'precision': 0.7, 'recall': 0.9130434782608695, 'f1': 0.7924528301886793, 'number': 92}                | {'precision': 0.9225248735932148, 'recall': 0.9867411025819958, 'f1': 0.9535530641490347, 'number': 5732} | 0.8286            | 0.8813         | 0.8542     | 0.8367           |
+| 0.4124        | 7.52  | 38000 | 0.4802          | {'precision': 0.6584084084084084, 'recall': 0.762498188668309, 'f1': 0.7066407036862955, 'number': 6901}  | {'precision': 0.7942951179374658, 'recall': 0.8174325392345038, 'f1': 0.8056977520587582, 'number': 8857} | {'precision': 0.9831375259156876, 'recall': 0.9894282932257615, 'f1': 0.9862728785357736, 'number': 7189} | {'precision': 0.6821705426356589, 'recall': 0.9565217391304348, 'f1': 0.7963800904977375, 'number': 92} | {'precision': 0.9221244705115673, 'recall': 0.9874389392882066, 'f1': 0.9536647009267059, 'number': 5732} | 0.8286            | 0.8815         | 0.8543     | 0.8362           |
+| 0.414         | 7.72  | 39000 | 0.4830          | {'precision': 0.6575084681972149, 'recall': 0.7594551514273293, 'f1': 0.7048144163528779, 'number': 6901} | {'precision': 0.7878721779910568, 'recall': 0.8156260584848143, 'f1': 0.8015089315433264, 'number': 8857} | {'precision': 0.9819235545743066, 'recall': 0.9898455974405341, 'f1': 0.9858686616791354, 'number': 7189} | {'precision': 0.6850393700787402, 'recall': 0.9456521739130435, 'f1': 0.7945205479452054, 'number': 92} | {'precision': 0.925229057591623, 'recall': 0.9865666434054431, 'f1': 0.9549138804457953, 'number': 5732}  | 0.8268            | 0.8802         | 0.8527     | 0.8366           |
+| 0.4212        | 7.92  | 40000 | 0.4803          | {'precision': 0.6614674319940105, 'recall': 0.7681495435444139, 'f1': 0.7108280254777071, 'number': 6901} | {'precision': 0.7921052631578948, 'recall': 0.8156260584848143, 'f1': 0.8036936084997496, 'number': 8857} | {'precision': 0.9846089850249584, 'recall': 0.9877590763666713, 'f1': 0.9861815151725575, 'number': 7189} | {'precision': 0.6854838709677419, 'recall': 0.9239130434782609, 'f1': 0.787037037037037, 'number': 92}  | {'precision': 0.9240444299248611, 'recall': 0.9869155617585486, 'f1': 0.9544457567065969, 'number': 5732} | 0.8292            | 0.8817         | 0.8547     | 0.8373           |
+| 0.3994        | 8.12  | 41000 | 0.4923          | {'precision': 0.6586345381526104, 'recall': 0.7604694971743226, 'f1': 0.7058981774161006, 'number': 6901} | {'precision': 0.7905782052683353, 'recall': 0.8166422039065147, 'f1': 0.8033988670443185, 'number': 8857} | {'precision': 0.9830016583747927, 'recall': 0.9894282932257615, 'f1': 0.9862045060658579, 'number': 7189} | {'precision': 0.6821705426356589, 'recall': 0.9565217391304348, 'f1': 0.7963800904977375, 'number': 92} | {'precision': 0.9232903541700669, 'recall': 0.9869155617585486, 'f1': 0.9540433426089888, 'number': 5732} | 0.8278            | 0.8807         | 0.8535     | 0.8355           |
+| 0.4052        | 8.32  | 42000 | 0.4866          | {'precision': 0.6577553593947036, 'recall': 0.7558324880452109, 'f1': 0.7033915447373744, 'number': 6901} | {'precision': 0.7920087575259989, 'recall': 0.8168680140002258, 'f1': 0.8042463317029791, 'number': 8857} | {'precision': 0.9823326432022084, 'recall': 0.9899846988454584, 'f1': 0.9861438270749618, 'number': 7189} | {'precision': 0.6846153846153846, 'recall': 0.967391304347826, 'f1': 0.8018018018018018, 'number': 92}  | {'precision': 0.9250531827851416, 'recall': 0.9862177250523377, 'f1': 0.9546567592670776, 'number': 5732} | 0.8285            | 0.8797         | 0.8533     | 0.8360           |
+| 0.4042        | 8.51  | 43000 | 0.4866          | {'precision': 0.663912546078556, 'recall': 0.7568468337922041, 'f1': 0.7073401950162513, 'number': 6901}  | {'precision': 0.7910823838737949, 'recall': 0.8152873433442475, 'f1': 0.8030025020850708, 'number': 8857} | {'precision': 0.980462300495322, 'recall': 0.991236611489776, 'f1': 0.9858200179843675, 'number': 7189}   | {'precision': 0.6796875, 'recall': 0.9456521739130435, 'f1': 0.7909090909090909, 'number': 92}          | {'precision': 0.921961550993809, 'recall': 0.9872644801116539, 'f1': 0.9534962089300758, 'number': 5732}  | 0.8292            | 0.8799         | 0.8538     | 0.8365           |
+| 0.4135        | 8.71  | 44000 | 0.4830          | {'precision': 0.664262212098203, 'recall': 0.7606144037096073, 'f1': 0.7091805715057758, 'number': 6901}  | {'precision': 0.7914884282110344, 'recall': 0.8147228181099695, 'f1': 0.802937576499388, 'number': 8857}  | {'precision': 0.9812594736116853, 'recall': 0.9905411044651551, 'f1': 0.985878443859892, 'number': 7189}  | {'precision': 0.6829268292682927, 'recall': 0.9130434782608695, 'f1': 0.7813953488372093, 'number': 92} | {'precision': 0.9230141901810471, 'recall': 0.9872644801116539, 'f1': 0.9540588384051252, 'number': 5732} | 0.8296            | 0.8804         | 0.8543     | 0.8368           |
+| 0.3958        | 8.91  | 45000 | 0.4859          | {'precision': 0.6619275289380976, 'recall': 0.7623532821330242, 'f1': 0.7085999057175567, 'number': 6901} | {'precision': 0.7909240021869874, 'recall': 0.8166422039065147, 'f1': 0.8035773802910788, 'number': 8857} | {'precision': 0.9831445150594087, 'recall': 0.9898455974405341, 'f1': 0.9864836764400082, 'number': 7189} | {'precision': 0.6910569105691057, 'recall': 0.9239130434782609, 'f1': 0.7906976744186046, 'number': 92} | {'precision': 0.9235543939888925, 'recall': 0.9863921842288904, 'f1': 0.9539395984477813, 'number': 5732} | 0.8290            | 0.8811         | 0.8543     | 0.8371           |
+| 0.3985        | 9.11  | 46000 | 0.4897          | {'precision': 0.6627135934710533, 'recall': 0.7530792638748007, 'f1': 0.7050125483280201, 'number': 6901} | {'precision': 0.7878688881629097, 'recall': 0.8168680140002258, 'f1': 0.8021064301552105, 'number': 8857} | {'precision': 0.9842192691029901, 'recall': 0.989010989010989, 'f1': 0.9866093110386457, 'number': 7189}  | {'precision': 0.6771653543307087, 'recall': 0.9347826086956522, 'f1': 0.7853881278538813, 'number': 92} | {'precision': 0.9269413889344935, 'recall': 0.984996510816469, 'f1': 0.9550875412331895, 'number': 5732}  | 0.8296            | 0.8785         | 0.8533     | 0.8352           |
+| 0.3944        | 9.31  | 47000 | 0.4890          | {'precision': 0.6661152818424736, 'recall': 0.7585857122156209, 'f1': 0.709349593495935, 'number': 6901}  | {'precision': 0.7902785363189514, 'recall': 0.8168680140002258, 'f1': 0.8033533200088829, 'number': 8857} | {'precision': 0.9839535205422604, 'recall': 0.9894282932257615, 'f1': 0.9866833125260092, 'number': 7189} | {'precision': 0.6854838709677419, 'recall': 0.9239130434782609, 'f1': 0.787037037037037, 'number': 92}  | {'precision': 0.9251923391717138, 'recall': 0.9860432658757851, 'f1': 0.954649100582721, 'number': 5732}  | 0.8308            | 0.8801         | 0.8547     | 0.8369           |
+| 0.3959        | 9.5   | 48000 | 0.4869          | {'precision': 0.6632755589047978, 'recall': 0.765251412838719, 'f1': 0.7106236964273699, 'number': 6901}  | {'precision': 0.7917122239230264, 'recall': 0.8175454442813593, 'f1': 0.8044214853080043, 'number': 8857} | {'precision': 0.9835476289229919, 'recall': 0.9895673946306858, 'f1': 0.9865483289418943, 'number': 7189} | {'precision': 0.688, 'recall': 0.9347826086956522, 'f1': 0.7926267281105991, 'number': 92}              | {'precision': 0.9240196078431373, 'recall': 0.9865666434054431, 'f1': 0.9542693216334796, 'number': 5732} | 0.8297            | 0.8820         | 0.8551     | 0.8369           |
+| 0.3979        | 9.7   | 49000 | 0.4878          | {'precision': 0.666244939271255, 'recall': 0.7630778148094479, 'f1': 0.7113812901046943, 'number': 6901}  | {'precision': 0.7906239755218009, 'recall': 0.8168680140002258, 'f1': 0.8035317636605953, 'number': 8857} | {'precision': 0.9835453539823009, 'recall': 0.9894282932257615, 'f1': 0.9864780528396089, 'number': 7189} | {'precision': 0.6829268292682927, 'recall': 0.9130434782608695, 'f1': 0.7813953488372093, 'number': 92} | {'precision': 0.9240071907174375, 'recall': 0.9863921842288904, 'f1': 0.9541810817652518, 'number': 5732} | 0.8304            | 0.8812         | 0.8550     | 0.8369           |
+| 0.4095        | 9.9   | 50000 | 0.4862          | {'precision': 0.665783497350492, 'recall': 0.7646717866975801, 'f1': 0.7118095366561004, 'number': 6901}  | {'precision': 0.7912472647702407, 'recall': 0.8165292988596591, 'f1': 0.80368950380619, 'number': 8857}   | {'precision': 0.9832734310201825, 'recall': 0.9894282932257615, 'f1': 0.9863412604867227, 'number': 7189} | {'precision': 0.6829268292682927, 'recall': 0.9130434782608695, 'f1': 0.7813953488372093, 'number': 92} | {'precision': 0.9244604316546763, 'recall': 0.9863921842288904, 'f1': 0.9544226873733964, 'number': 5732} | 0.8304            | 0.8814         | 0.8552     | 0.8373           |
+
+
+### Framework versions
+
+- Transformers 4.34.1
+- Pytorch 2.1.0+cu118
+- Datasets 2.14.5
+- Tokenizers 0.14.1

all_results.json

@@ -0,0 +1,41 @@
+{
+    "epoch": 10.0,
+    "eval_COMMENT": {
+        "f1": 0.7114860480207656,
+        "number": 7171,
+        "precision": 0.6653720111664037,
+        "recall": 0.7644679960953842
+    },
+    "eval_NAME": {
+        "f1": 0.8086344637849849,
+        "number": 9305,
+        "precision": 0.7963734889537307,
+        "recall": 0.8212788823213326
+    },
+    "eval_QTY": {
+        "f1": 0.9862540673351484,
+        "number": 7495,
+        "precision": 0.981755684822845,
+        "recall": 0.9907938625750501
+    },
+    "eval_RANGE_END": {
+        "f1": 0.7443946188340806,
+        "number": 90,
+        "precision": 0.6240601503759399,
+        "recall": 0.9222222222222223
+    },
+    "eval_UNIT": {
+        "f1": 0.95358036651328,
+        "number": 5989,
+        "precision": 0.923100969052829,
+        "recall": 0.9861412589747871
+    },
+    "eval_loss": 0.4705313444137573,
+    "eval_overall_accuracy": 0.8349855394073885,
+    "eval_overall_f1": 0.8563730235559858,
+    "eval_overall_precision": 0.8311619166927654,
+    "eval_overall_recall": 0.8831613976705491,
+    "eval_runtime": 8.4988,
+    "eval_samples_per_second": 1053.557,
+    "eval_steps_per_second": 32.946
+}

config.json

@@ -0,0 +1,51 @@
+{
+  "_name_or_path": "napsternxg/gte-small-L3-ingredient-v2",
+  "architectures": [
+    "PretrainedTaggerModel"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 384,
+  "id2label": {
+    "0": "O",
+    "1": "B-COMMENT",
+    "2": "I-COMMENT",
+    "3": "B-NAME",
+    "4": "I-NAME",
+    "5": "B-RANGE_END",
+    "6": "I-RANGE_END",
+    "7": "B-QTY",
+    "8": "I-QTY",
+    "9": "B-UNIT",
+    "10": "I-UNIT"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 1536,
+  "label2id": {
+    "B-COMMENT": 1,
+    "B-NAME": 3,
+    "B-QTY": 7,
+    "B-RANGE_END": 5,
+    "B-UNIT": 9,
+    "I-COMMENT": 2,
+    "I-NAME": 4,
+    "I-QTY": 8,
+    "I-RANGE_END": 6,
+    "I-UNIT": 10,
+    "O": 0
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 3,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.34.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2395ae8a9abeffa945966cd12ef3f2d1b87f42814c0ff25275981260c3d1ed2b
+size 69004115

special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

test_results.json

@@ -0,0 +1,41 @@
+{
+    "epoch": 10.0,
+    "eval_COMMENT": {
+        "f1": 0.7114860480207656,
+        "number": 7171,
+        "precision": 0.6653720111664037,
+        "recall": 0.7644679960953842
+    },
+    "eval_NAME": {
+        "f1": 0.8086344637849849,
+        "number": 9305,
+        "precision": 0.7963734889537307,
+        "recall": 0.8212788823213326
+    },
+    "eval_QTY": {
+        "f1": 0.9862540673351484,
+        "number": 7495,
+        "precision": 0.981755684822845,
+        "recall": 0.9907938625750501
+    },
+    "eval_RANGE_END": {
+        "f1": 0.7443946188340806,
+        "number": 90,
+        "precision": 0.6240601503759399,
+        "recall": 0.9222222222222223
+    },
+    "eval_UNIT": {
+        "f1": 0.95358036651328,
+        "number": 5989,
+        "precision": 0.923100969052829,
+        "recall": 0.9861412589747871
+    },
+    "eval_loss": 0.4705313444137573,
+    "eval_overall_accuracy": 0.8349855394073885,
+    "eval_overall_f1": 0.8563730235559858,
+    "eval_overall_precision": 0.8311619166927654,
+    "eval_overall_recall": 0.8831613976705491,
+    "eval_runtime": 8.4988,
+    "eval_samples_per_second": 1053.557,
+    "eval_steps_per_second": 32.946
+}

tokenizer_config.json

@@ -0,0 +1,64 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "max_length": 128,
+  "model_max_length": 1000000000000000019884624838656,
+  "never_split": null,
+  "pad_to_multiple_of": null,
+  "pad_token": "[PAD]",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "sep_token": "[SEP]",
+  "stride": 0,
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first",
+  "unk_token": "[UNK]"
+}

train_results.json

@@ -0,0 +1,41 @@
+{
+    "epoch": 10.0,
+    "eval_COMMENT": {
+        "f1": 0.7388161975793072,
+        "number": 129716,
+        "precision": 0.694144754676064,
+        "recall": 0.7896327361312405
+    },
+    "eval_NAME": {
+        "f1": 0.8224766455255967,
+        "number": 167638,
+        "precision": 0.809525735315358,
+        "recall": 0.8358486739283456
+    },
+    "eval_QTY": {
+        "f1": 0.9868797185369786,
+        "number": 135780,
+        "precision": 0.9822001750802452,
+        "recall": 0.9916040653999116
+    },
+    "eval_RANGE_END": {
+        "f1": 0.7616517749335909,
+        "number": 1682,
+        "precision": 0.6413176087840585,
+        "recall": 0.9375743162901308
+    },
+    "eval_UNIT": {
+        "f1": 0.9564042026259744,
+        "number": 108832,
+        "precision": 0.926564721085505,
+        "recall": 0.9882295648338724
+    },
+    "eval_loss": 0.4124242663383484,
+    "eval_overall_accuracy": 0.855277993837768,
+    "eval_overall_f1": 0.868490442544095,
+    "eval_overall_precision": 0.8439132826504565,
+    "eval_overall_recall": 0.894542056624875,
+    "eval_runtime": 130.0825,
+    "eval_samples_per_second": 1242.311,
+    "eval_steps_per_second": 38.829
+}

trainer_state.json

@@ -0,0 +1,2734 @@
+{
+  "best_metric": 0.474032998085022,
+  "best_model_checkpoint": "nyt_ingredients-tagger-gte-small-L3-ingredient-v2/checkpoint-24000",
+  "epoch": 10.0,
+  "eval_steps": 1000,
+  "global_step": 50510,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.9505048505246485e-05,
+      "loss": 1.1293,
+      "step": 500
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.9010097010492975e-05,
+      "loss": 0.6931,
+      "step": 1000
+    },
+    {
+      "epoch": 0.2,
+      "eval_COMMENT": {
+        "f1": 0.5770606987183658,
+        "number": 6901,
+        "precision": 0.5236127508854782,
+        "recall": 0.6426604839878278
+      },
+      "eval_NAME": {
+        "f1": 0.7875208913649026,
+        "number": 8857,
+        "precision": 0.7773012207192346,
+        "recall": 0.7980128711753416
+      },
+      "eval_QTY": {
+        "f1": 0.9780634343998882,
+        "number": 7189,
+        "precision": 0.9824561403508771,
+        "recall": 0.9737098344693281
+      },
+      "eval_RANGE_END": {
+        "f1": 0.0,
+        "number": 92,
+        "precision": 0.0,
+        "recall": 0.0
+      },
+      "eval_UNIT": {
+        "f1": 0.9434346865975879,
+        "number": 5732,
+        "precision": 0.9192320423700762,
+        "recall": 0.9689462665736218
+      },
+      "eval_loss": 0.6230219006538391,
+      "eval_overall_accuracy": 0.7985813450669499,
+      "eval_overall_f1": 0.8086250651249558,
+      "eval_overall_precision": 0.7828506345590628,
+      "eval_overall_recall": 0.8361544610892913,
+      "eval_runtime": 7.9128,
+      "eval_samples_per_second": 1074.967,
+      "eval_steps_per_second": 33.616,
+      "step": 1000
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.851514551573946e-05,
+      "loss": 0.6087,
+      "step": 1500
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.802019402098595e-05,
+      "loss": 0.5826,
+      "step": 2000
+    },
+    {
+      "epoch": 0.4,
+      "eval_COMMENT": {
+        "f1": 0.602454780361757,
+        "number": 6901,
+        "precision": 0.5435365427205968,
+        "recall": 0.6756991740327489
+      },
+      "eval_NAME": {
+        "f1": 0.7907440212577502,
+        "number": 8857,
+        "precision": 0.7757141305528402,
+        "recall": 0.8063678446426555
+      },
+      "eval_QTY": {
+        "f1": 0.9754149163280765,
+        "number": 7189,
+        "precision": 0.9659028914348063,
+        "recall": 0.9851161496731117
+      },
+      "eval_RANGE_END": {
+        "f1": 0.6567164179104478,
+        "number": 92,
+        "precision": 0.6055045871559633,
+        "recall": 0.717391304347826
+      },
+      "eval_UNIT": {
+        "f1": 0.9462474645030426,
+        "number": 5732,
+        "precision": 0.9177049180327869,
+        "recall": 0.9766224703419399
+      },
+      "eval_loss": 0.5530511736869812,
+      "eval_overall_accuracy": 0.8078557068618898,
+      "eval_overall_f1": 0.8170321807714068,
+      "eval_overall_precision": 0.7837009608325087,
+      "eval_overall_recall": 0.8533245281707275,
+      "eval_runtime": 6.3157,
+      "eval_samples_per_second": 1346.806,
+      "eval_steps_per_second": 42.117,
+      "step": 2000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.752524252623243e-05,
+      "loss": 0.5569,
+      "step": 2500
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.703029103147892e-05,
+      "loss": 0.5435,
+      "step": 3000
+    },
+    {
+      "epoch": 0.59,
+      "eval_COMMENT": {
+        "f1": 0.6313494946826079,
+        "number": 6901,
+        "precision": 0.5801165331391115,
+        "recall": 0.6925083321257789
+      },
+      "eval_NAME": {
+        "f1": 0.7944305207463102,
+        "number": 8857,
+        "precision": 0.7839085513299626,
+        "recall": 0.8052387941740996
+      },
+      "eval_QTY": {
+        "f1": 0.9791782343142699,
+        "number": 7189,
+        "precision": 0.9771436487048067,
+        "recall": 0.9812213103352344
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7109004739336493,
+        "number": 92,
+        "precision": 0.6302521008403361,
+        "recall": 0.8152173913043478
+      },
+      "eval_UNIT": {
+        "f1": 0.9472093999160722,
+        "number": 5732,
+        "precision": 0.9126637554585153,
+        "recall": 0.9844731332868109
+      },
+      "eval_loss": 0.5327398180961609,
+      "eval_overall_accuracy": 0.8150438114023442,
+      "eval_overall_f1": 0.827899644462333,
+      "eval_overall_precision": 0.7999157403506497,
+      "eval_overall_recall": 0.8579124813179938,
+      "eval_runtime": 8.1233,
+      "eval_samples_per_second": 1047.112,
+      "eval_steps_per_second": 32.745,
+      "step": 3000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.65353395367254e-05,
+      "loss": 0.5324,
+      "step": 3500
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.6040388041971886e-05,
+      "loss": 0.5333,
+      "step": 4000
+    },
+    {
+      "epoch": 0.79,
+      "eval_COMMENT": {
+        "f1": 0.6446886446886447,
+        "number": 6901,
+        "precision": 0.5875760104924288,
+        "recall": 0.7140994058832053
+      },
+      "eval_NAME": {
+        "f1": 0.7953999888349244,
+        "number": 8857,
+        "precision": 0.786660777385159,
+        "recall": 0.8043355537992548
+      },
+      "eval_QTY": {
+        "f1": 0.9809583594894328,
+        "number": 7189,
+        "precision": 0.9837716843872412,
+        "recall": 0.9781610794269022
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7555555555555556,
+        "number": 92,
+        "precision": 0.6390977443609023,
+        "recall": 0.9239130434782609
+      },
+      "eval_UNIT": {
+        "f1": 0.9506497918967128,
+        "number": 5732,
+        "precision": 0.9263366992219831,
+        "recall": 0.9762735519888346
+      },
+      "eval_loss": 0.5212323665618896,
+      "eval_overall_accuracy": 0.8183059591093578,
+      "eval_overall_f1": 0.831933619994625,
+      "eval_overall_precision": 0.8049731838127743,
+      "eval_overall_recall": 0.8607625734246289,
+      "eval_runtime": 8.6379,
+      "eval_samples_per_second": 984.732,
+      "eval_steps_per_second": 30.795,
+      "step": 4000
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.5545436547218376e-05,
+      "loss": 0.5285,
+      "step": 4500
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.505048505246486e-05,
+      "loss": 0.5199,
+      "step": 5000
+    },
+    {
+      "epoch": 0.99,
+      "eval_COMMENT": {
+        "f1": 0.6402057097646205,
+        "number": 6901,
+        "precision": 0.5873457536898137,
+        "recall": 0.7035212288074192
+      },
+      "eval_NAME": {
+        "f1": 0.8005563282336579,
+        "number": 8857,
+        "precision": 0.7890984865101996,
+        "recall": 0.812351812126002
+      },
+      "eval_QTY": {
+        "f1": 0.9803540437348144,
+        "number": 7189,
+        "precision": 0.978519955654102,
+        "recall": 0.9821950201697037
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7747747747747747,
+        "number": 92,
+        "precision": 0.6615384615384615,
+        "recall": 0.9347826086956522
+      },
+      "eval_UNIT": {
+        "f1": 0.9499704716105627,
+        "number": 5732,
+        "precision": 0.9197843489625878,
+        "recall": 0.9822051639916259
+      },
+      "eval_loss": 0.5120549201965332,
+      "eval_overall_accuracy": 0.8185525167848879,
+      "eval_overall_f1": 0.8328133910301567,
+      "eval_overall_precision": 0.804738906356358,
+      "eval_overall_recall": 0.8629175211150116,
+      "eval_runtime": 6.4358,
+      "eval_samples_per_second": 1321.664,
+      "eval_steps_per_second": 41.331,
+      "step": 5000
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 4.455553355771135e-05,
+      "loss": 0.5127,
+      "step": 5500
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 4.406058206295783e-05,
+      "loss": 0.4983,
+      "step": 6000
+    },
+    {
+      "epoch": 1.19,
+      "eval_COMMENT": {
+        "f1": 0.6476600322754169,
+        "number": 6901,
+        "precision": 0.6041901894367081,
+        "recall": 0.6978698739313143
+      },
+      "eval_NAME": {
+        "f1": 0.7935644937586686,
+        "number": 8857,
+        "precision": 0.7801047120418848,
+        "recall": 0.8074968951112115
+      },
+      "eval_QTY": {
+        "f1": 0.9824293353705119,
+        "number": 7189,
+        "precision": 0.980998613037448,
+        "recall": 0.983864237028794
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7601809954751132,
+        "number": 92,
+        "precision": 0.6511627906976745,
+        "recall": 0.9130434782608695
+      },
+      "eval_UNIT": {
+        "f1": 0.9500843170320405,
+        "number": 5732,
+        "precision": 0.9193864229765013,
+        "recall": 0.9829030006978368
+      },
+      "eval_loss": 0.5088106393814087,
+      "eval_overall_accuracy": 0.8192163259113151,
+      "eval_overall_f1": 0.8339592771611903,
+      "eval_overall_precision": 0.8089590276416389,
+      "eval_overall_recall": 0.8605540300997532,
+      "eval_runtime": 8.5928,
+      "eval_samples_per_second": 989.902,
+      "eval_steps_per_second": 30.956,
+      "step": 6000
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 4.356563056820432e-05,
+      "loss": 0.5102,
+      "step": 6500
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 4.3070679073450804e-05,
+      "loss": 0.5031,
+      "step": 7000
+    },
+    {
+      "epoch": 1.39,
+      "eval_COMMENT": {
+        "f1": 0.6483065113460373,
+        "number": 6901,
+        "precision": 0.6055345911949686,
+        "recall": 0.6975800608607449
+      },
+      "eval_NAME": {
+        "f1": 0.797360248447205,
+        "number": 8857,
+        "precision": 0.7835422343324251,
+        "recall": 0.8116743818448685
+      },
+      "eval_QTY": {
+        "f1": 0.9815493055075669,
+        "number": 7189,
+        "precision": 0.9752815160670145,
+        "recall": 0.9878981777715955
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7706422018348623,
+        "number": 92,
+        "precision": 0.6666666666666666,
+        "recall": 0.9130434782608695
+      },
+      "eval_UNIT": {
+        "f1": 0.9498936622713738,
+        "number": 5732,
+        "precision": 0.9269467043001827,
+        "recall": 0.9740055826936497
+      },
+      "eval_loss": 0.5030972361564636,
+      "eval_overall_accuracy": 0.8199370329628646,
+      "eval_overall_f1": 0.8351003758828189,
+      "eval_overall_precision": 0.8107082078806126,
+      "eval_overall_recall": 0.8610058739703174,
+      "eval_runtime": 7.9063,
+      "eval_samples_per_second": 1075.848,
+      "eval_steps_per_second": 33.644,
+      "step": 7000
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 4.257572757869729e-05,
+      "loss": 0.5027,
+      "step": 7500
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 4.208077608394378e-05,
+      "loss": 0.492,
+      "step": 8000
+    },
+    {
+      "epoch": 1.58,
+      "eval_COMMENT": {
+        "f1": 0.6769068494956819,
+        "number": 6901,
+        "precision": 0.6209482341557814,
+        "recall": 0.743950152151862
+      },
+      "eval_NAME": {
+        "f1": 0.7926012072880323,
+        "number": 8857,
+        "precision": 0.7778260869565218,
+        "recall": 0.8079485152986339
+      },
+      "eval_QTY": {
+        "f1": 0.9821801475706529,
+        "number": 7189,
+        "precision": 0.983001254005852,
+        "recall": 0.9813604117401585
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7685589519650656,
+        "number": 92,
+        "precision": 0.6423357664233577,
+        "recall": 0.9565217391304348
+      },
+      "eval_UNIT": {
+        "f1": 0.9496828752642706,
+        "number": 5732,
+        "precision": 0.9215493188905302,
+        "recall": 0.9795882763433357
+      },
+      "eval_loss": 0.5041355490684509,
+      "eval_overall_accuracy": 0.8246216287979365,
+      "eval_overall_f1": 0.8398886765248299,
+      "eval_overall_precision": 0.8112712550607287,
+      "eval_overall_recall": 0.8705988669146015,
+      "eval_runtime": 9.0062,
+      "eval_samples_per_second": 944.457,
+      "eval_steps_per_second": 29.535,
+      "step": 8000
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 4.158582458919026e-05,
+      "loss": 0.4939,
+      "step": 8500
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 4.109087309443675e-05,
+      "loss": 0.4817,
+      "step": 9000
+    },
+    {
+      "epoch": 1.78,
+      "eval_COMMENT": {
+        "f1": 0.6750676255195619,
+        "number": 6901,
+        "precision": 0.6196705426356589,
+        "recall": 0.7413418345167367
+      },
+      "eval_NAME": {
+        "f1": 0.8006897702620014,
+        "number": 8857,
+        "precision": 0.7891447368421053,
+        "recall": 0.8125776222197132
+      },
+      "eval_QTY": {
+        "f1": 0.9833679833679833,
+        "number": 7189,
+        "precision": 0.9798370390829996,
+        "recall": 0.9869244679371262
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7792207792207793,
+        "number": 92,
+        "precision": 0.6474820143884892,
+        "recall": 0.9782608695652174
+      },
+      "eval_UNIT": {
+        "f1": 0.9505083606419629,
+        "number": 5732,
+        "precision": 0.9168422758956071,
+        "recall": 0.9867411025819958
+      },
+      "eval_loss": 0.4901449382305145,
+      "eval_overall_accuracy": 0.8292872586579676,
+      "eval_overall_f1": 0.8427365317609219,
+      "eval_overall_precision": 0.8133872271624899,
+      "eval_overall_recall": 0.8742831323207396,
+      "eval_runtime": 9.1291,
+      "eval_samples_per_second": 931.746,
+      "eval_steps_per_second": 29.138,
+      "step": 9000
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 4.059592159968323e-05,
+      "loss": 0.4903,
+      "step": 9500
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 4.010097010492972e-05,
+      "loss": 0.4905,
+      "step": 10000
+    },
+    {
+      "epoch": 1.98,
+      "eval_COMMENT": {
+        "f1": 0.6778359049069598,
+        "number": 6901,
+        "precision": 0.6241463414634146,
+        "recall": 0.7416316475873062
+      },
+      "eval_NAME": {
+        "f1": 0.8002232142857143,
+        "number": 8857,
+        "precision": 0.7911287653095002,
+        "recall": 0.8095291859546122
+      },
+      "eval_QTY": {
+        "f1": 0.9842749791260784,
+        "number": 7189,
+        "precision": 0.9846860643185299,
+        "recall": 0.983864237028794
+      },
+      "eval_RANGE_END": {
+        "f1": 0.789237668161435,
+        "number": 92,
+        "precision": 0.6717557251908397,
+        "recall": 0.9565217391304348
+      },
+      "eval_UNIT": {
+        "f1": 0.9502270052127123,
+        "number": 5732,
+        "precision": 0.9170723790976956,
+        "recall": 0.9858688066992324
+      },
+      "eval_loss": 0.4891791045665741,
+      "eval_overall_accuracy": 0.8287751773318667,
+      "eval_overall_f1": 0.8435557049235421,
+      "eval_overall_precision": 0.8165522625980025,
+      "eval_overall_recall": 0.8724062423968579,
+      "eval_runtime": 6.9634,
+      "eval_samples_per_second": 1221.536,
+      "eval_steps_per_second": 38.2,
+      "step": 10000
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 3.9606018610176205e-05,
+      "loss": 0.4804,
+      "step": 10500
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 3.911106711542269e-05,
+      "loss": 0.4757,
+      "step": 11000
+    },
+    {
+      "epoch": 2.18,
+      "eval_COMMENT": {
+        "f1": 0.6788539520042544,
+        "number": 6901,
+        "precision": 0.6271186440677966,
+        "recall": 0.7398927691638892
+      },
+      "eval_NAME": {
+        "f1": 0.802429240026744,
+        "number": 8857,
+        "precision": 0.7921020789792103,
+        "recall": 0.8130292424071356
+      },
+      "eval_QTY": {
+        "f1": 0.9835042971998891,
+        "number": 7189,
+        "precision": 0.9801077496891836,
+        "recall": 0.9869244679371262
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7758620689655172,
+        "number": 92,
+        "precision": 0.6428571428571429,
+        "recall": 0.9782608695652174
+      },
+      "eval_UNIT": {
+        "f1": 0.9515842839036756,
+        "number": 5732,
+        "precision": 0.9226609864001311,
+        "recall": 0.9823796231681786
+      },
+      "eval_loss": 0.4953509271144867,
+      "eval_overall_accuracy": 0.8274285931039714,
+      "eval_overall_f1": 0.844669333960932,
+      "eval_overall_precision": 0.8179391176949373,
+      "eval_overall_recall": 0.8732056584755483,
+      "eval_runtime": 8.0512,
+      "eval_samples_per_second": 1056.487,
+      "eval_steps_per_second": 33.039,
+      "step": 11000
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 3.861611562066917e-05,
+      "loss": 0.4856,
+      "step": 11500
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 3.812116412591566e-05,
+      "loss": 0.4691,
+      "step": 12000
+    },
+    {
+      "epoch": 2.38,
+      "eval_COMMENT": {
+        "f1": 0.6842557693579316,
+        "number": 6901,
+        "precision": 0.6292872780345414,
+        "recall": 0.7497464135632517
+      },
+      "eval_NAME": {
+        "f1": 0.803944509443423,
+        "number": 8857,
+        "precision": 0.7935547734271887,
+        "recall": 0.8146099130631139
+      },
+      "eval_QTY": {
+        "f1": 0.984621778886118,
+        "number": 7189,
+        "precision": 0.9806816613771215,
+        "recall": 0.9885936847962165
+      },
+      "eval_RANGE_END": {
+        "f1": 0.780701754385965,
+        "number": 92,
+        "precision": 0.6544117647058824,
+        "recall": 0.967391304347826
+      },
+      "eval_UNIT": {
+        "f1": 0.9528556944913822,
+        "number": 5732,
+        "precision": 0.9238204456094364,
+        "recall": 0.9837752965806001
+      },
+      "eval_loss": 0.4902682304382324,
+      "eval_overall_accuracy": 0.8301407275348026,
+      "eval_overall_f1": 0.84684079769019,
+      "eval_overall_precision": 0.818934450180189,
+      "eval_overall_recall": 0.8767161377776233,
+      "eval_runtime": 8.9123,
+      "eval_samples_per_second": 954.413,
+      "eval_steps_per_second": 29.846,
+      "step": 12000
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 3.762621263116215e-05,
+      "loss": 0.4806,
+      "step": 12500
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 3.7131261136408633e-05,
+      "loss": 0.4765,
+      "step": 13000
+    },
+    {
+      "epoch": 2.57,
+      "eval_COMMENT": {
+        "f1": 0.6893633899649634,
+        "number": 6901,
+        "precision": 0.6338439095550693,
+        "recall": 0.7555426749746413
+      },
+      "eval_NAME": {
+        "f1": 0.805656475322788,
+        "number": 8857,
+        "precision": 0.7977640026566305,
+        "recall": 0.8137066726882691
+      },
+      "eval_QTY": {
+        "f1": 0.9841977027497391,
+        "number": 7189,
+        "precision": 0.9850891861761427,
+        "recall": 0.9833078314090973
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7777777777777777,
+        "number": 92,
+        "precision": 0.6774193548387096,
+        "recall": 0.9130434782608695
+      },
+      "eval_UNIT": {
+        "f1": 0.9518021440027011,
+        "number": 5732,
+        "precision": 0.9219950940310712,
+        "recall": 0.9836008374040475
+      },
+      "eval_loss": 0.4845636188983917,
+      "eval_overall_accuracy": 0.8307097067860258,
+      "eval_overall_f1": 0.8482320088820106,
+      "eval_overall_precision": 0.8219070904645477,
+      "eval_overall_recall": 0.8762990511278718,
+      "eval_runtime": 6.7092,
+      "eval_samples_per_second": 1267.814,
+      "eval_steps_per_second": 39.647,
+      "step": 13000
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 3.663630964165512e-05,
+      "loss": 0.4715,
+      "step": 13500
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 3.6141358146901606e-05,
+      "loss": 0.4747,
+      "step": 14000
+    },
+    {
+      "epoch": 2.77,
+      "eval_COMMENT": {
+        "f1": 0.6842317515795134,
+        "number": 6901,
+        "precision": 0.6380844929171368,
+        "recall": 0.7375742645993334
+      },
+      "eval_NAME": {
+        "f1": 0.8026484170700496,
+        "number": 8857,
+        "precision": 0.791246160596753,
+        "recall": 0.8143841029694028
+      },
+      "eval_QTY": {
+        "f1": 0.9847090569431951,
+        "number": 7189,
+        "precision": 0.9796255506607929,
+        "recall": 0.9898455974405341
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7878787878787878,
+        "number": 92,
+        "precision": 0.6546762589928058,
+        "recall": 0.9891304347826086
+      },
+      "eval_UNIT": {
+        "f1": 0.9519481612387444,
+        "number": 5732,
+        "precision": 0.9195252804422045,
+        "recall": 0.9867411025819958
+      },
+      "eval_loss": 0.4844052493572235,
+      "eval_overall_accuracy": 0.8314304138375754,
+      "eval_overall_f1": 0.8470833754081255,
+      "eval_overall_precision": 0.8211570463666917,
+      "eval_overall_recall": 0.8747002189704911,
+      "eval_runtime": 8.8562,
+      "eval_samples_per_second": 960.458,
+      "eval_steps_per_second": 30.035,
+      "step": 14000
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 3.564640665214809e-05,
+      "loss": 0.4673,
+      "step": 14500
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 3.515145515739457e-05,
+      "loss": 0.4654,
+      "step": 15000
+    },
+    {
+      "epoch": 2.97,
+      "eval_COMMENT": {
+        "f1": 0.6951928862739855,
+        "number": 6901,
+        "precision": 0.6453574975173784,
+        "recall": 0.7533690769453703
+      },
+      "eval_NAME": {
+        "f1": 0.8027195720017833,
+        "number": 8857,
+        "precision": 0.7925608011444921,
+        "recall": 0.8131421474539912
+      },
+      "eval_QTY": {
+        "f1": 0.9852237252861602,
+        "number": 7189,
+        "precision": 0.9827013562136728,
+        "recall": 0.9877590763666713
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7802690582959643,
+        "number": 92,
+        "precision": 0.6641221374045801,
+        "recall": 0.9456521739130435
+      },
+      "eval_UNIT": {
+        "f1": 0.9533801506049582,
+        "number": 5732,
+        "precision": 0.9255791030064071,
+        "recall": 0.9829030006978368
+      },
+      "eval_loss": 0.48082008957862854,
+      "eval_overall_accuracy": 0.8331752835413269,
+      "eval_overall_f1": 0.8498601704909194,
+      "eval_overall_precision": 0.8246313793441659,
+      "eval_overall_recall": 0.8766813805568107,
+      "eval_runtime": 7.2911,
+      "eval_samples_per_second": 1166.621,
+      "eval_steps_per_second": 36.483,
+      "step": 15000
+    },
+    {
+      "epoch": 3.07,
+      "learning_rate": 3.465650366264107e-05,
+      "loss": 0.4573,
+      "step": 15500
+    },
+    {
+      "epoch": 3.17,
+      "learning_rate": 3.416155216788755e-05,
+      "loss": 0.4586,
+      "step": 16000
+    },
+    {
+      "epoch": 3.17,
+      "eval_COMMENT": {
+        "f1": 0.6976900801400768,
+        "number": 6901,
+        "precision": 0.6517362858580775,
+        "recall": 0.7506158527749601
+      },
+      "eval_NAME": {
+        "f1": 0.8010911317708622,
+        "number": 8857,
+        "precision": 0.7901383703052932,
+        "recall": 0.812351812126002
+      },
+      "eval_QTY": {
+        "f1": 0.9851262538913871,
+        "number": 7189,
+        "precision": 0.979906413432425,
+        "recall": 0.990402003060231
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7802690582959643,
+        "number": 92,
+        "precision": 0.6641221374045801,
+        "recall": 0.9456521739130435
+      },
+      "eval_UNIT": {
+        "f1": 0.9516034003871727,
+        "number": 5732,
+        "precision": 0.9193364774760123,
+        "recall": 0.9862177250523377
+      },
+      "eval_loss": 0.48265689611434937,
+      "eval_overall_accuracy": 0.8316959374881463,
+      "eval_overall_f1": 0.8500783210658401,
+      "eval_overall_precision": 0.824673202614379,
+      "eval_overall_recall": 0.8770984672065621,
+      "eval_runtime": 9.1659,
+      "eval_samples_per_second": 928.008,
+      "eval_steps_per_second": 29.021,
+      "step": 16000
+    },
+    {
+      "epoch": 3.27,
+      "learning_rate": 3.3666600673134034e-05,
+      "loss": 0.4591,
+      "step": 16500
+    },
+    {
+      "epoch": 3.37,
+      "learning_rate": 3.317164917838052e-05,
+      "loss": 0.4612,
+      "step": 17000
+    },
+    {
+      "epoch": 3.37,
+      "eval_COMMENT": {
+        "f1": 0.6966127636920211,
+        "number": 6901,
+        "precision": 0.6441053408811224,
+        "recall": 0.7584408056803362
+      },
+      "eval_NAME": {
+        "f1": 0.7990913120567377,
+        "number": 8857,
+        "precision": 0.7845718637797846,
+        "recall": 0.8141582928756915
+      },
+      "eval_QTY": {
+        "f1": 0.9846325012168833,
+        "number": 7189,
+        "precision": 0.9844271412680756,
+        "recall": 0.9848379468632633
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7802690582959643,
+        "number": 92,
+        "precision": 0.6641221374045801,
+        "recall": 0.9456521739130435
+      },
+      "eval_UNIT": {
+        "f1": 0.9502140878179833,
+        "number": 5732,
+        "precision": 0.9158439877002751,
+        "recall": 0.9872644801116539
+      },
+      "eval_loss": 0.4868798851966858,
+      "eval_overall_accuracy": 0.8307666047111482,
+      "eval_overall_f1": 0.8481624433629804,
+      "eval_overall_precision": 0.8199811804406373,
+      "eval_overall_recall": 0.8783497271558166,
+      "eval_runtime": 7.3722,
+      "eval_samples_per_second": 1153.794,
+      "eval_steps_per_second": 36.081,
+      "step": 17000
+    },
+    {
+      "epoch": 3.46,
+      "learning_rate": 3.267669768362701e-05,
+      "loss": 0.4592,
+      "step": 17500
+    },
+    {
+      "epoch": 3.56,
+      "learning_rate": 3.218174618887349e-05,
+      "loss": 0.4546,
+      "step": 18000
+    },
+    {
+      "epoch": 3.56,
+      "eval_COMMENT": {
+        "f1": 0.6903326332904275,
+        "number": 6901,
+        "precision": 0.6482188295165394,
+        "recall": 0.7382987972757571
+      },
+      "eval_NAME": {
+        "f1": 0.8029488387561666,
+        "number": 8857,
+        "precision": 0.788654181184669,
+        "recall": 0.8177712543750706
+      },
+      "eval_QTY": {
+        "f1": 0.9853083853083853,
+        "number": 7189,
+        "precision": 0.9817704736914791,
+        "recall": 0.9888718876060648
+      },
+      "eval_RANGE_END": {
+        "f1": 0.787037037037037,
+        "number": 92,
+        "precision": 0.6854838709677419,
+        "recall": 0.9239130434782609
+      },
+      "eval_UNIT": {
+        "f1": 0.9534019922336655,
+        "number": 5732,
+        "precision": 0.9236179260713118,
+        "recall": 0.9851709699930217
+      },
+      "eval_loss": 0.480613112449646,
+      "eval_overall_accuracy": 0.8314683457876569,
+      "eval_overall_f1": 0.8492933517725234,
+      "eval_overall_precision": 0.8249189136061331,
+      "eval_overall_recall": 0.8751520628410552,
+      "eval_runtime": 8.0402,
+      "eval_samples_per_second": 1057.931,
+      "eval_steps_per_second": 33.084,
+      "step": 18000
+    },
+    {
+      "epoch": 3.66,
+      "learning_rate": 3.168679469411997e-05,
+      "loss": 0.4609,
+      "step": 18500
+    },
+    {
+      "epoch": 3.76,
+      "learning_rate": 3.119184319936647e-05,
+      "loss": 0.4471,
+      "step": 19000
+    },
+    {
+      "epoch": 3.76,
+      "eval_COMMENT": {
+        "f1": 0.6993611924407772,
+        "number": 6901,
+        "precision": 0.6466100652147164,
+        "recall": 0.7614838429213158
+      },
+      "eval_NAME": {
+        "f1": 0.8037861915367484,
+        "number": 8857,
+        "precision": 0.7929254092057564,
+        "recall": 0.8149486282036807
+      },
+      "eval_QTY": {
+        "f1": 0.9852052510939778,
+        "number": 7189,
+        "precision": 0.983906770255272,
+        "recall": 0.9865071637223536
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7927927927927929,
+        "number": 92,
+        "precision": 0.676923076923077,
+        "recall": 0.9565217391304348
+      },
+      "eval_UNIT": {
+        "f1": 0.9520276536548351,
+        "number": 5732,
+        "precision": 0.9211943220753793,
+        "recall": 0.984996510816469
+      },
+      "eval_loss": 0.4822600483894348,
+      "eval_overall_accuracy": 0.8327769980654706,
+      "eval_overall_f1": 0.8508441514764242,
+      "eval_overall_precision": 0.8241521972831222,
+      "eval_overall_recall": 0.87932292933857,
+      "eval_runtime": 8.7949,
+      "eval_samples_per_second": 967.151,
+      "eval_steps_per_second": 30.245,
+      "step": 19000
+    },
+    {
+      "epoch": 3.86,
+      "learning_rate": 3.069689170461295e-05,
+      "loss": 0.4581,
+      "step": 19500
+    },
+    {
+      "epoch": 3.96,
+      "learning_rate": 3.0201940209859435e-05,
+      "loss": 0.4542,
+      "step": 20000
+    },
+    {
+      "epoch": 3.96,
+      "eval_COMMENT": {
+        "f1": 0.6950766529344229,
+        "number": 6901,
+        "precision": 0.6508980521123198,
+        "recall": 0.7456890305752789
+      },
+      "eval_NAME": {
+        "f1": 0.7992666259236624,
+        "number": 8857,
+        "precision": 0.7868081382629621,
+        "recall": 0.8121260020322908
+      },
+      "eval_QTY": {
+        "f1": 0.9853604384930272,
+        "number": 7189,
+        "precision": 0.9829734219269103,
+        "recall": 0.9877590763666713
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7873303167420815,
+        "number": 92,
+        "precision": 0.6744186046511628,
+        "recall": 0.9456521739130435
+      },
+      "eval_UNIT": {
+        "f1": 0.9524210526315788,
+        "number": 5732,
+        "precision": 0.920559986977047,
+        "recall": 0.9865666434054431
+      },
+      "eval_loss": 0.47578319907188416,
+      "eval_overall_accuracy": 0.832568372340022,
+      "eval_overall_f1": 0.8490938211245047,
+      "eval_overall_precision": 0.8244499738082766,
+      "eval_overall_recall": 0.875256334503493,
+      "eval_runtime": 6.7831,
+      "eval_samples_per_second": 1253.998,
+      "eval_steps_per_second": 39.215,
+      "step": 20000
+    },
+    {
+      "epoch": 4.06,
+      "learning_rate": 2.970698871510592e-05,
+      "loss": 0.4454,
+      "step": 20500
+    },
+    {
+      "epoch": 4.16,
+      "learning_rate": 2.9212037220352405e-05,
+      "loss": 0.4452,
+      "step": 21000
+    },
+    {
+      "epoch": 4.16,
+      "eval_COMMENT": {
+        "f1": 0.6982899022801302,
+        "number": 6901,
+        "precision": 0.6566687938736439,
+        "recall": 0.7455441240399941
+      },
+      "eval_NAME": {
+        "f1": 0.7992468294844105,
+        "number": 8857,
+        "precision": 0.7843478260869565,
+        "recall": 0.8147228181099695
+      },
+      "eval_QTY": {
+        "f1": 0.9847455276660656,
+        "number": 7189,
+        "precision": 0.981750311074243,
+        "recall": 0.9877590763666713
+      },
+      "eval_RANGE_END": {
+        "f1": 0.787037037037037,
+        "number": 92,
+        "precision": 0.6854838709677419,
+        "recall": 0.9239130434782609
+      },
+      "eval_UNIT": {
+        "f1": 0.9537733457280487,
+        "number": 5732,
+        "precision": 0.9249303392886412,
+        "recall": 0.9844731332868109
+      },
+      "eval_loss": 0.48325347900390625,
+      "eval_overall_accuracy": 0.8338959905928763,
+      "eval_overall_f1": 0.8500944924406048,
+      "eval_overall_precision": 0.8260912340537172,
+      "eval_overall_recall": 0.8755343922699941,
+      "eval_runtime": 9.0461,
+      "eval_samples_per_second": 940.291,
+      "eval_steps_per_second": 29.405,
+      "step": 21000
+    },
+    {
+      "epoch": 4.26,
+      "learning_rate": 2.871708572559889e-05,
+      "loss": 0.4217,
+      "step": 21500
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 2.8222134230845377e-05,
+      "loss": 0.4412,
+      "step": 22000
+    },
+    {
+      "epoch": 4.36,
+      "eval_COMMENT": {
+        "f1": 0.6974386700006758,
+        "number": 6901,
+        "precision": 0.6534954407294833,
+        "recall": 0.7477177220692653
+      },
+      "eval_NAME": {
+        "f1": 0.8008429926238145,
+        "number": 8857,
+        "precision": 0.7870067582297798,
+        "recall": 0.8151744382973919
+      },
+      "eval_QTY": {
+        "f1": 0.9858392336526447,
+        "number": 7189,
+        "precision": 0.9839268394069558,
+        "recall": 0.9877590763666713
+      },
+      "eval_RANGE_END": {
+        "f1": 0.789237668161435,
+        "number": 92,
+        "precision": 0.6717557251908397,
+        "recall": 0.9565217391304348
+      },
+      "eval_UNIT": {
+        "f1": 0.9540462916033113,
+        "number": 5732,
+        "precision": 0.9248280379954144,
+        "recall": 0.9851709699930217
+      },
+      "eval_loss": 0.4888293445110321,
+      "eval_overall_accuracy": 0.8325494063649812,
+      "eval_overall_f1": 0.8505270258875117,
+      "eval_overall_precision": 0.8261040492727034,
+      "eval_overall_recall": 0.8764380800111223,
+      "eval_runtime": 6.5585,
+      "eval_samples_per_second": 1296.944,
+      "eval_steps_per_second": 40.558,
+      "step": 22000
+    },
+    {
+      "epoch": 4.45,
+      "learning_rate": 2.7727182736091867e-05,
+      "loss": 0.4481,
+      "step": 22500
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 2.7232231241338353e-05,
+      "loss": 0.4423,
+      "step": 23000
+    },
+    {
+      "epoch": 4.55,
+      "eval_COMMENT": {
+        "f1": 0.6981985021253626,
+        "number": 6901,
+        "precision": 0.6532828282828282,
+        "recall": 0.7497464135632517
+      },
+      "eval_NAME": {
+        "f1": 0.8019725177304966,
+        "number": 8857,
+        "precision": 0.7874007180937874,
+        "recall": 0.817093824093937
+      },
+      "eval_QTY": {
+        "f1": 0.9857043719639139,
+        "number": 7189,
+        "precision": 0.9835202880487467,
+        "recall": 0.9878981777715955
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7999999999999999,
+        "number": 92,
+        "precision": 0.6766917293233082,
+        "recall": 0.9782608695652174
+      },
+      "eval_UNIT": {
+        "f1": 0.9548452562151193,
+        "number": 5732,
+        "precision": 0.9264850672792911,
+        "recall": 0.984996510816469
+      },
+      "eval_loss": 0.4827423095703125,
+      "eval_overall_accuracy": 0.8337821947426317,
+      "eval_overall_f1": 0.8511377043654138,
+      "eval_overall_precision": 0.8262377695605223,
+      "eval_overall_recall": 0.8775850682979389,
+      "eval_runtime": 8.9572,
+      "eval_samples_per_second": 949.629,
+      "eval_steps_per_second": 29.697,
+      "step": 23000
+    },
+    {
+      "epoch": 4.65,
+      "learning_rate": 2.6737279746584836e-05,
+      "loss": 0.4519,
+      "step": 23500
+    },
+    {
+      "epoch": 4.75,
+      "learning_rate": 2.6242328251831323e-05,
+      "loss": 0.4507,
+      "step": 24000
+    },
+    {
+      "epoch": 4.75,
+      "eval_COMMENT": {
+        "f1": 0.69833923223523,
+        "number": 6901,
+        "precision": 0.6584520600693108,
+        "recall": 0.7433705260107231
+      },
+      "eval_NAME": {
+        "f1": 0.8048291977300545,
+        "number": 8857,
+        "precision": 0.7933530766699572,
+        "recall": 0.8166422039065147
+      },
+      "eval_QTY": {
+        "f1": 0.9856135011758197,
+        "number": 7189,
+        "precision": 0.9801898472967395,
+        "recall": 0.9910975100848518
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7837837837837838,
+        "number": 92,
+        "precision": 0.6692307692307692,
+        "recall": 0.9456521739130435
+      },
+      "eval_UNIT": {
+        "f1": 0.9537942664418213,
+        "number": 5732,
+        "precision": 0.922976501305483,
+        "recall": 0.9867411025819958
+      },
+      "eval_loss": 0.474032998085022,
+      "eval_overall_accuracy": 0.8347115275196297,
+      "eval_overall_f1": 0.8523122656487518,
+      "eval_overall_precision": 0.8290126499096435,
+      "eval_overall_recall": 0.8769594383233117,
+      "eval_runtime": 7.5421,
+      "eval_samples_per_second": 1127.8,
+      "eval_steps_per_second": 35.269,
+      "step": 24000
+    },
+    {
+      "epoch": 4.85,
+      "learning_rate": 2.5747376757077806e-05,
+      "loss": 0.4377,
+      "step": 24500
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 2.5252425262324292e-05,
+      "loss": 0.4452,
+      "step": 25000
+    },
+    {
+      "epoch": 4.95,
+      "eval_COMMENT": {
+        "f1": 0.7040598290598291,
+        "number": 6901,
+        "precision": 0.6528792569659443,
+        "recall": 0.7639472540211564
+      },
+      "eval_NAME": {
+        "f1": 0.8012646292084975,
+        "number": 8857,
+        "precision": 0.7875054513737462,
+        "recall": 0.8155131534379587
+      },
+      "eval_QTY": {
+        "f1": 0.9855775897933713,
+        "number": 7189,
+        "precision": 0.9825798423890502,
+        "recall": 0.9885936847962165
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7727272727272728,
+        "number": 92,
+        "precision": 0.6640625,
+        "recall": 0.9239130434782609
+      },
+      "eval_UNIT": {
+        "f1": 0.953449148254343,
+        "number": 5732,
+        "precision": 0.9227881162259223,
+        "recall": 0.9862177250523377
+      },
+      "eval_loss": 0.4791179597377777,
+      "eval_overall_accuracy": 0.8352236088457308,
+      "eval_overall_f1": 0.8516931350306697,
+      "eval_overall_precision": 0.8244940456823062,
+      "eval_overall_recall": 0.8807479753918876,
+      "eval_runtime": 6.5842,
+      "eval_samples_per_second": 1291.88,
+      "eval_steps_per_second": 40.4,
+      "step": 25000
+    },
+    {
+      "epoch": 5.05,
+      "learning_rate": 2.4757473767570778e-05,
+      "loss": 0.4442,
+      "step": 25500
+    },
+    {
+      "epoch": 5.15,
+      "learning_rate": 2.4262522272817265e-05,
+      "loss": 0.4274,
+      "step": 26000
+    },
+    {
+      "epoch": 5.15,
+      "eval_COMMENT": {
+        "f1": 0.7096947333109694,
+        "number": 6901,
+        "precision": 0.6607946026986506,
+        "recall": 0.766410665120997
+      },
+      "eval_NAME": {
+        "f1": 0.8056639536180176,
+        "number": 8857,
+        "precision": 0.795727342803656,
+        "recall": 0.8158518685785254
+      },
+      "eval_QTY": {
+        "f1": 0.9859017987360232,
+        "number": 7189,
+        "precision": 0.9844660194174757,
+        "recall": 0.9873417721518988
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7962962962962964,
+        "number": 92,
+        "precision": 0.6935483870967742,
+        "recall": 0.9347826086956522
+      },
+      "eval_UNIT": {
+        "f1": 0.9532647207693606,
+        "number": 5732,
+        "precision": 0.9229010127409344,
+        "recall": 0.9856943475226797
+      },
+      "eval_loss": 0.48070788383483887,
+      "eval_overall_accuracy": 0.836058111747525,
+      "eval_overall_f1": 0.854768006474238,
+      "eval_overall_precision": 0.8299990177138928,
+      "eval_overall_recall": 0.8810607903792013,
+      "eval_runtime": 9.3326,
+      "eval_samples_per_second": 911.433,
+      "eval_steps_per_second": 28.502,
+      "step": 26000
+    },
+    {
+      "epoch": 5.25,
+      "learning_rate": 2.376757077806375e-05,
+      "loss": 0.432,
+      "step": 26500
+    },
+    {
+      "epoch": 5.35,
+      "learning_rate": 2.3272619283310237e-05,
+      "loss": 0.4312,
+      "step": 27000
+    },
+    {
+      "epoch": 5.35,
+      "eval_COMMENT": {
+        "f1": 0.7056545830829829,
+        "number": 6901,
+        "precision": 0.6542461005199307,
+        "recall": 0.7658310389798579
+      },
+      "eval_NAME": {
+        "f1": 0.8019135562107137,
+        "number": 8857,
+        "precision": 0.7903508771929825,
+        "recall": 0.8138195777351248
+      },
+      "eval_QTY": {
+        "f1": 0.9852421533984619,
+        "number": 7189,
+        "precision": 0.981501932633904,
+        "recall": 0.989010989010989
+      },
+      "eval_RANGE_END": {
+        "f1": 0.789237668161435,
+        "number": 92,
+        "precision": 0.6717557251908397,
+        "recall": 0.9565217391304348
+      },
+      "eval_UNIT": {
+        "f1": 0.9502391541495343,
+        "number": 5732,
+        "precision": 0.9154405820533549,
+        "recall": 0.9877878576413119
+      },
+      "eval_loss": 0.48605257272720337,
+      "eval_overall_accuracy": 0.83349770511702,
+      "eval_overall_f1": 0.8517865242150884,
+      "eval_overall_precision": 0.8242733597763183,
+      "eval_overall_recall": 0.8811998192624517,
+      "eval_runtime": 7.0035,
+      "eval_samples_per_second": 1214.536,
+      "eval_steps_per_second": 37.981,
+      "step": 27000
+    },
+    {
+      "epoch": 5.44,
+      "learning_rate": 2.2777667788556724e-05,
+      "loss": 0.4321,
+      "step": 27500
+    },
+    {
+      "epoch": 5.54,
+      "learning_rate": 2.2282716293803206e-05,
+      "loss": 0.4315,
+      "step": 28000
+    },
+    {
+      "epoch": 5.54,
+      "eval_COMMENT": {
+        "f1": 0.7045745469244761,
+        "number": 6901,
+        "precision": 0.6583983883152859,
+        "recall": 0.7577162730039124
+      },
+      "eval_NAME": {
+        "f1": 0.8038757099899767,
+        "number": 8857,
+        "precision": 0.7930996593780903,
+        "recall": 0.8149486282036807
+      },
+      "eval_QTY": {
+        "f1": 0.9861322978782415,
+        "number": 7189,
+        "precision": 0.9831328632655882,
+        "recall": 0.9891500904159132
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7909090909090909,
+        "number": 92,
+        "precision": 0.6796875,
+        "recall": 0.9456521739130435
+      },
+      "eval_UNIT": {
+        "f1": 0.9528698872243729,
+        "number": 5732,
+        "precision": 0.9204878048780488,
+        "recall": 0.9876133984647593
+      },
+      "eval_loss": 0.481955349445343,
+      "eval_overall_accuracy": 0.8341235822933657,
+      "eval_overall_f1": 0.8531310577328276,
+      "eval_overall_precision": 0.8282385285069058,
+      "eval_overall_recall": 0.8795662298842585,
+      "eval_runtime": 8.4264,
+      "eval_samples_per_second": 1009.444,
+      "eval_steps_per_second": 31.567,
+      "step": 28000
+    },
+    {
+      "epoch": 5.64,
+      "learning_rate": 2.1787764799049696e-05,
+      "loss": 0.4328,
+      "step": 28500
+    },
+    {
+      "epoch": 5.74,
+      "learning_rate": 2.129281330429618e-05,
+      "loss": 0.4316,
+      "step": 29000
+    },
+    {
+      "epoch": 5.74,
+      "eval_COMMENT": {
+        "f1": 0.7066999527697188,
+        "number": 6901,
+        "precision": 0.6612373737373738,
+        "recall": 0.7588755252861904
+      },
+      "eval_NAME": {
+        "f1": 0.8088874039425326,
+        "number": 8857,
+        "precision": 0.7980441709702231,
+        "recall": 0.8200293553121825
+      },
+      "eval_QTY": {
+        "f1": 0.9862747816442535,
+        "number": 7189,
+        "precision": 0.9830040071852978,
+        "recall": 0.9895673946306858
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7741935483870968,
+        "number": 92,
+        "precision": 0.672,
+        "recall": 0.9130434782608695
+      },
+      "eval_UNIT": {
+        "f1": 0.955117910573916,
+        "number": 5732,
+        "precision": 0.9263813739957371,
+        "recall": 0.9856943475226797
+      },
+      "eval_loss": 0.4761992394924164,
+      "eval_overall_accuracy": 0.836987444524523,
+      "eval_overall_f1": 0.8555853712048336,
+      "eval_overall_precision": 0.8315727314480678,
+      "eval_overall_recall": 0.8810260331583887,
+      "eval_runtime": 8.3244,
+      "eval_samples_per_second": 1021.82,
+      "eval_steps_per_second": 31.954,
+      "step": 29000
+    },
+    {
+      "epoch": 5.84,
+      "learning_rate": 2.0797861809542665e-05,
+      "loss": 0.4429,
+      "step": 29500
+    },
+    {
+      "epoch": 5.94,
+      "learning_rate": 2.0302910314789152e-05,
+      "loss": 0.4278,
+      "step": 30000
+    },
+    {
+      "epoch": 5.94,
+      "eval_COMMENT": {
+        "f1": 0.7081487447979596,
+        "number": 6901,
+        "precision": 0.6596223583843942,
+        "recall": 0.7643819736270105
+      },
+      "eval_NAME": {
+        "f1": 0.8039357385068653,
+        "number": 8857,
+        "precision": 0.7918309242225142,
+        "recall": 0.8164163938128034
+      },
+      "eval_QTY": {
+        "f1": 0.9861265260821309,
+        "number": 7189,
+        "precision": 0.9835339698353397,
+        "recall": 0.9887327862011406
+      },
+      "eval_RANGE_END": {
+        "f1": 0.783410138248848,
+        "number": 92,
+        "precision": 0.68,
+        "recall": 0.9239130434782609
+      },
+      "eval_UNIT": {
+        "f1": 0.9540695710908478,
+        "number": 5732,
+        "precision": 0.9244109947643979,
+        "recall": 0.9856943475226797
+      },
+      "eval_loss": 0.4822809398174286,
+      "eval_overall_accuracy": 0.8347684254447522,
+      "eval_overall_f1": 0.8540192709386161,
+      "eval_overall_precision": 0.8285882391396725,
+      "eval_overall_recall": 0.8810607903792013,
+      "eval_runtime": 6.5281,
+      "eval_samples_per_second": 1302.987,
+      "eval_steps_per_second": 40.747,
+      "step": 30000
+    },
+    {
+      "epoch": 6.04,
+      "learning_rate": 1.9807958820035638e-05,
+      "loss": 0.4251,
+      "step": 30500
+    },
+    {
+      "epoch": 6.14,
+      "learning_rate": 1.9313007325282124e-05,
+      "loss": 0.4211,
+      "step": 31000
+    },
+    {
+      "epoch": 6.14,
+      "eval_COMMENT": {
+        "f1": 0.705458989679522,
+        "number": 6901,
+        "precision": 0.6637281206081512,
+        "recall": 0.7527894508042313
+      },
+      "eval_NAME": {
+        "f1": 0.7998453978245265,
+        "number": 8857,
+        "precision": 0.7826885671061162,
+        "recall": 0.8177712543750706
+      },
+      "eval_QTY": {
+        "f1": 0.9858901646147462,
+        "number": 7189,
+        "precision": 0.9804649883065071,
+        "recall": 0.9913757128947003
+      },
+      "eval_RANGE_END": {
+        "f1": 0.787037037037037,
+        "number": 92,
+        "precision": 0.6854838709677419,
+        "recall": 0.9239130434782609
+      },
+      "eval_UNIT": {
+        "f1": 0.9548103724976772,
+        "number": 5732,
+        "precision": 0.925495333224169,
+        "recall": 0.9860432658757851
+      },
+      "eval_loss": 0.47935572266578674,
+      "eval_overall_accuracy": 0.8356218943215871,
+      "eval_overall_f1": 0.8526081682167409,
+      "eval_overall_precision": 0.8273764755894183,
+      "eval_overall_recall": 0.8794272010010079,
+      "eval_runtime": 9.2091,
+      "eval_samples_per_second": 923.653,
+      "eval_steps_per_second": 28.885,
+      "step": 31000
+    },
+    {
+      "epoch": 6.24,
+      "learning_rate": 1.8818055830528607e-05,
+      "loss": 0.4113,
+      "step": 31500
+    },
+    {
+      "epoch": 6.34,
+      "learning_rate": 1.8323104335775097e-05,
+      "loss": 0.4231,
+      "step": 32000
+    },
+    {
+      "epoch": 6.34,
+      "eval_COMMENT": {
+        "f1": 0.7081927385612093,
+        "number": 6901,
+        "precision": 0.6627510420613869,
+        "recall": 0.7603245906390378
+      },
+      "eval_NAME": {
+        "f1": 0.8008871638480732,
+        "number": 8857,
+        "precision": 0.7868816735672259,
+        "recall": 0.815400248391103
+      },
+      "eval_QTY": {
+        "f1": 0.9866055937261433,
+        "number": 7189,
+        "precision": 0.9844875346260388,
+        "recall": 0.9887327862011406
+      },
+      "eval_RANGE_END": {
+        "f1": 0.8018433179723502,
+        "number": 92,
+        "precision": 0.696,
+        "recall": 0.9456521739130435
+      },
+      "eval_UNIT": {
+        "f1": 0.9542042675212956,
+        "number": 5732,
+        "precision": 0.9235918367346939,
+        "recall": 0.9869155617585486
+      },
+      "eval_loss": 0.4756552577018738,
+      "eval_overall_accuracy": 0.8364374312483405,
+      "eval_overall_f1": 0.8534784953485236,
+      "eval_overall_precision": 0.8284312121707835,
+      "eval_overall_recall": 0.8800875881964478,
+      "eval_runtime": 6.7948,
+      "eval_samples_per_second": 1251.831,
+      "eval_steps_per_second": 39.147,
+      "step": 32000
+    },
+    {
+      "epoch": 6.43,
+      "learning_rate": 1.782815284102158e-05,
+      "loss": 0.4209,
+      "step": 32500
+    },
+    {
+      "epoch": 6.53,
+      "learning_rate": 1.7333201346268066e-05,
+      "loss": 0.4302,
+      "step": 33000
+    },
+    {
+      "epoch": 6.53,
+      "eval_COMMENT": {
+        "f1": 0.706057529610829,
+        "number": 6901,
+        "precision": 0.6624333248666497,
+        "recall": 0.7558324880452109
+      },
+      "eval_NAME": {
+        "f1": 0.8062360801781736,
+        "number": 8857,
+        "precision": 0.7953421948808085,
+        "recall": 0.8174325392345038
+      },
+      "eval_QTY": {
+        "f1": 0.986130374479889,
+        "number": 7189,
+        "precision": 0.9832664914949523,
+        "recall": 0.989010989010989
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7766990291262136,
+        "number": 92,
+        "precision": 0.7017543859649122,
+        "recall": 0.8695652173913043
+      },
+      "eval_UNIT": {
+        "f1": 0.9544303797468354,
+        "number": 5732,
+        "precision": 0.9243216737495914,
+        "recall": 0.9865666434054431
+      },
+      "eval_loss": 0.4785662591457367,
+      "eval_overall_accuracy": 0.8354891324963016,
+      "eval_overall_f1": 0.8546047187178056,
+      "eval_overall_precision": 0.8311760840998685,
+      "eval_overall_recall": 0.8793924437801953,
+      "eval_runtime": 8.3624,
+      "eval_samples_per_second": 1017.166,
+      "eval_steps_per_second": 31.809,
+      "step": 33000
+    },
+    {
+      "epoch": 6.63,
+      "learning_rate": 1.6838249851514553e-05,
+      "loss": 0.4309,
+      "step": 33500
+    },
+    {
+      "epoch": 6.73,
+      "learning_rate": 1.634329835676104e-05,
+      "loss": 0.4223,
+      "step": 34000
+    },
+    {
+      "epoch": 6.73,
+      "eval_COMMENT": {
+        "f1": 0.703981612925032,
+        "number": 6901,
+        "precision": 0.6597820577800304,
+        "recall": 0.7545283292276481
+      },
+      "eval_NAME": {
+        "f1": 0.8044022011005503,
+        "number": 8857,
+        "precision": 0.7922049485439019,
+        "recall": 0.8169809190470814
+      },
+      "eval_QTY": {
+        "f1": 0.986130374479889,
+        "number": 7189,
+        "precision": 0.9832664914949523,
+        "recall": 0.989010989010989
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7981651376146789,
+        "number": 92,
+        "precision": 0.6904761904761905,
+        "recall": 0.9456521739130435
+      },
+      "eval_UNIT": {
+        "f1": 0.9541965415436525,
+        "number": 5732,
+        "precision": 0.923730197615548,
+        "recall": 0.9867411025819958
+      },
+      "eval_loss": 0.4795403480529785,
+      "eval_overall_accuracy": 0.836304669423055,
+      "eval_overall_f1": 0.8534844880813807,
+      "eval_overall_precision": 0.8292139251294828,
+      "eval_overall_recall": 0.8792186576761322,
+      "eval_runtime": 8.6534,
+      "eval_samples_per_second": 982.967,
+      "eval_steps_per_second": 30.739,
+      "step": 34000
+    },
+    {
+      "epoch": 6.83,
+      "learning_rate": 1.5848346862007525e-05,
+      "loss": 0.4119,
+      "step": 34500
+    },
+    {
+      "epoch": 6.93,
+      "learning_rate": 1.535339536725401e-05,
+      "loss": 0.4184,
+      "step": 35000
+    },
+    {
+      "epoch": 6.93,
+      "eval_COMMENT": {
+        "f1": 0.7059539095762654,
+        "number": 6901,
+        "precision": 0.6614741641337386,
+        "recall": 0.7568468337922041
+      },
+      "eval_NAME": {
+        "f1": 0.8020405899966729,
+        "number": 8857,
+        "precision": 0.788057099269914,
+        "recall": 0.8165292988596591
+      },
+      "eval_QTY": {
+        "f1": 0.9865650969529086,
+        "number": 7189,
+        "precision": 0.9823472624465591,
+        "recall": 0.9908193072750034
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7963800904977375,
+        "number": 92,
+        "precision": 0.6821705426356589,
+        "recall": 0.9565217391304348
+      },
+      "eval_UNIT": {
+        "f1": 0.9530710253601821,
+        "number": 5732,
+        "precision": 0.9216229428059313,
+        "recall": 0.9867411025819958
+      },
+      "eval_loss": 0.4803193211555481,
+      "eval_overall_accuracy": 0.8355839623715056,
+      "eval_overall_f1": 0.8531527433837031,
+      "eval_overall_precision": 0.8277868584504741,
+      "eval_overall_recall": 0.8801223454172604,
+      "eval_runtime": 7.0311,
+      "eval_samples_per_second": 1209.766,
+      "eval_steps_per_second": 37.832,
+      "step": 35000
+    },
+    {
+      "epoch": 7.03,
+      "learning_rate": 1.4858443872500496e-05,
+      "loss": 0.4189,
+      "step": 35500
+    },
+    {
+      "epoch": 7.13,
+      "learning_rate": 1.4363492377746981e-05,
+      "loss": 0.4071,
+      "step": 36000
+    },
+    {
+      "epoch": 7.13,
+      "eval_COMMENT": {
+        "f1": 0.7068977142471848,
+        "number": 6901,
+        "precision": 0.6610340479192938,
+        "recall": 0.7596000579626141
+      },
+      "eval_NAME": {
+        "f1": 0.8058371393561323,
+        "number": 8857,
+        "precision": 0.7952072111685171,
+        "recall": 0.8167551089533702
+      },
+      "eval_QTY": {
+        "f1": 0.9864247125640672,
+        "number": 7189,
+        "precision": 0.9823423920540765,
+        "recall": 0.9905411044651551
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7981651376146789,
+        "number": 92,
+        "precision": 0.6904761904761905,
+        "recall": 0.9456521739130435
+      },
+      "eval_UNIT": {
+        "f1": 0.9542881284326151,
+        "number": 5732,
+        "precision": 0.9252826478780928,
+        "recall": 0.9851709699930217
+      },
+      "eval_loss": 0.48823681473731995,
+      "eval_overall_accuracy": 0.8358115540719948,
+      "eval_overall_f1": 0.8546798029556651,
+      "eval_overall_precision": 0.8303884609080479,
+      "eval_overall_recall": 0.880435160404574,
+      "eval_runtime": 9.3911,
+      "eval_samples_per_second": 905.75,
+      "eval_steps_per_second": 28.325,
+      "step": 36000
+    },
+    {
+      "epoch": 7.23,
+      "learning_rate": 1.3868540882993467e-05,
+      "loss": 0.4066,
+      "step": 36500
+    },
+    {
+      "epoch": 7.33,
+      "learning_rate": 1.3373589388239954e-05,
+      "loss": 0.4093,
+      "step": 37000
+    },
+    {
+      "epoch": 7.33,
+      "eval_COMMENT": {
+        "f1": 0.7065691471515801,
+        "number": 6901,
+        "precision": 0.6579605098725319,
+        "recall": 0.7629329082741632
+      },
+      "eval_NAME": {
+        "f1": 0.8055648302726766,
+        "number": 8857,
+        "precision": 0.794249972566663,
+        "recall": 0.8172067291407926
+      },
+      "eval_QTY": {
+        "f1": 0.9863412604867227,
+        "number": 7189,
+        "precision": 0.9832734310201825,
+        "recall": 0.9894282932257615
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7924528301886793,
+        "number": 92,
+        "precision": 0.7,
+        "recall": 0.9130434782608695
+      },
+      "eval_UNIT": {
+        "f1": 0.9535530641490347,
+        "number": 5732,
+        "precision": 0.9225248735932148,
+        "recall": 0.9867411025819958
+      },
+      "eval_loss": 0.48563364148139954,
+      "eval_overall_accuracy": 0.8367219208739521,
+      "eval_overall_f1": 0.8541543851375251,
+      "eval_overall_precision": 0.8286274509803921,
+      "eval_overall_recall": 0.8813040909248896,
+      "eval_runtime": 6.6204,
+      "eval_samples_per_second": 1284.823,
+      "eval_steps_per_second": 40.179,
+      "step": 37000
+    },
+    {
+      "epoch": 7.42,
+      "learning_rate": 1.287863789348644e-05,
+      "loss": 0.4197,
+      "step": 37500
+    },
+    {
+      "epoch": 7.52,
+      "learning_rate": 1.2383686398732925e-05,
+      "loss": 0.4124,
+      "step": 38000
+    },
+    {
+      "epoch": 7.52,
+      "eval_COMMENT": {
+        "f1": 0.7066407036862955,
+        "number": 6901,
+        "precision": 0.6584084084084084,
+        "recall": 0.762498188668309
+      },
+      "eval_NAME": {
+        "f1": 0.8056977520587582,
+        "number": 8857,
+        "precision": 0.7942951179374658,
+        "recall": 0.8174325392345038
+      },
+      "eval_QTY": {
+        "f1": 0.9862728785357736,
+        "number": 7189,
+        "precision": 0.9831375259156876,
+        "recall": 0.9894282932257615
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7963800904977375,
+        "number": 92,
+        "precision": 0.6821705426356589,
+        "recall": 0.9565217391304348
+      },
+      "eval_UNIT": {
+        "f1": 0.9536647009267059,
+        "number": 5732,
+        "precision": 0.9221244705115673,
+        "recall": 0.9874389392882066
+      },
+      "eval_loss": 0.4802015423774719,
+      "eval_overall_accuracy": 0.8361908735728104,
+      "eval_overall_f1": 0.8542606938363086,
+      "eval_overall_precision": 0.8286124995916234,
+      "eval_overall_recall": 0.881547391470578,
+      "eval_runtime": 9.142,
+      "eval_samples_per_second": 930.428,
+      "eval_steps_per_second": 29.096,
+      "step": 38000
+    },
+    {
+      "epoch": 7.62,
+      "learning_rate": 1.1888734903979411e-05,
+      "loss": 0.4071,
+      "step": 38500
+    },
+    {
+      "epoch": 7.72,
+      "learning_rate": 1.1393783409225896e-05,
+      "loss": 0.414,
+      "step": 39000
+    },
+    {
+      "epoch": 7.72,
+      "eval_COMMENT": {
+        "f1": 0.7048144163528779,
+        "number": 6901,
+        "precision": 0.6575084681972149,
+        "recall": 0.7594551514273293
+      },
+      "eval_NAME": {
+        "f1": 0.8015089315433264,
+        "number": 8857,
+        "precision": 0.7878721779910568,
+        "recall": 0.8156260584848143
+      },
+      "eval_QTY": {
+        "f1": 0.9858686616791354,
+        "number": 7189,
+        "precision": 0.9819235545743066,
+        "recall": 0.9898455974405341
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7945205479452054,
+        "number": 92,
+        "precision": 0.6850393700787402,
+        "recall": 0.9456521739130435
+      },
+      "eval_UNIT": {
+        "f1": 0.9549138804457953,
+        "number": 5732,
+        "precision": 0.925229057591623,
+        "recall": 0.9865666434054431
+      },
+      "eval_loss": 0.483047753572464,
+      "eval_overall_accuracy": 0.8365701930736259,
+      "eval_overall_f1": 0.8526693267336735,
+      "eval_overall_precision": 0.8268464703193366,
+      "eval_overall_recall": 0.880157102638073,
+      "eval_runtime": 7.5994,
+      "eval_samples_per_second": 1119.299,
+      "eval_steps_per_second": 35.003,
+      "step": 39000
+    },
+    {
+      "epoch": 7.82,
+      "learning_rate": 1.0898831914472382e-05,
+      "loss": 0.4196,
+      "step": 39500
+    },
+    {
+      "epoch": 7.92,
+      "learning_rate": 1.0403880419718868e-05,
+      "loss": 0.4212,
+      "step": 40000
+    },
+    {
+      "epoch": 7.92,
+      "eval_COMMENT": {
+        "f1": 0.7108280254777071,
+        "number": 6901,
+        "precision": 0.6614674319940105,
+        "recall": 0.7681495435444139
+      },
+      "eval_NAME": {
+        "f1": 0.8036936084997496,
+        "number": 8857,
+        "precision": 0.7921052631578948,
+        "recall": 0.8156260584848143
+      },
+      "eval_QTY": {
+        "f1": 0.9861815151725575,
+        "number": 7189,
+        "precision": 0.9846089850249584,
+        "recall": 0.9877590763666713
+      },
+      "eval_RANGE_END": {
+        "f1": 0.787037037037037,
+        "number": 92,
+        "precision": 0.6854838709677419,
+        "recall": 0.9239130434782609
+      },
+      "eval_UNIT": {
+        "f1": 0.9544457567065969,
+        "number": 5732,
+        "precision": 0.9240444299248611,
+        "recall": 0.9869155617585486
+      },
+      "eval_loss": 0.4802730679512024,
+      "eval_overall_accuracy": 0.8373477980502978,
+      "eval_overall_f1": 0.8546737867021544,
+      "eval_overall_precision": 0.8292364016736402,
+      "eval_overall_recall": 0.8817211775746411,
+      "eval_runtime": 8.1983,
+      "eval_samples_per_second": 1037.537,
+      "eval_steps_per_second": 32.446,
+      "step": 40000
+    },
+    {
+      "epoch": 8.02,
+      "learning_rate": 9.908928924965353e-06,
+      "loss": 0.4135,
+      "step": 40500
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 9.41397743021184e-06,
+      "loss": 0.3994,
+      "step": 41000
+    },
+    {
+      "epoch": 8.12,
+      "eval_COMMENT": {
+        "f1": 0.7058981774161006,
+        "number": 6901,
+        "precision": 0.6586345381526104,
+        "recall": 0.7604694971743226
+      },
+      "eval_NAME": {
+        "f1": 0.8033988670443185,
+        "number": 8857,
+        "precision": 0.7905782052683353,
+        "recall": 0.8166422039065147
+      },
+      "eval_QTY": {
+        "f1": 0.9862045060658579,
+        "number": 7189,
+        "precision": 0.9830016583747927,
+        "recall": 0.9894282932257615
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7963800904977375,
+        "number": 92,
+        "precision": 0.6821705426356589,
+        "recall": 0.9565217391304348
+      },
+      "eval_UNIT": {
+        "f1": 0.9540433426089888,
+        "number": 5732,
+        "precision": 0.9232903541700669,
+        "recall": 0.9869155617585486
+      },
+      "eval_loss": 0.4922930598258972,
+      "eval_overall_accuracy": 0.8354701665212608,
+      "eval_overall_f1": 0.8534523408555069,
+      "eval_overall_precision": 0.8278284164788134,
+      "eval_overall_recall": 0.8807132181710751,
+      "eval_runtime": 7.4213,
+      "eval_samples_per_second": 1146.156,
+      "eval_steps_per_second": 35.843,
+      "step": 41000
+    },
+    {
+      "epoch": 8.22,
+      "learning_rate": 8.919025935458326e-06,
+      "loss": 0.4167,
+      "step": 41500
+    },
+    {
+      "epoch": 8.32,
+      "learning_rate": 8.424074440704812e-06,
+      "loss": 0.4052,
+      "step": 42000
+    },
+    {
+      "epoch": 8.32,
+      "eval_COMMENT": {
+        "f1": 0.7033915447373744,
+        "number": 6901,
+        "precision": 0.6577553593947036,
+        "recall": 0.7558324880452109
+      },
+      "eval_NAME": {
+        "f1": 0.8042463317029791,
+        "number": 8857,
+        "precision": 0.7920087575259989,
+        "recall": 0.8168680140002258
+      },
+      "eval_QTY": {
+        "f1": 0.9861438270749618,
+        "number": 7189,
+        "precision": 0.9823326432022084,
+        "recall": 0.9899846988454584
+      },
+      "eval_RANGE_END": {
+        "f1": 0.8018018018018018,
+        "number": 92,
+        "precision": 0.6846153846153846,
+        "recall": 0.967391304347826
+      },
+      "eval_UNIT": {
+        "f1": 0.9546567592670776,
+        "number": 5732,
+        "precision": 0.9250531827851416,
+        "recall": 0.9862177250523377
+      },
+      "eval_loss": 0.4865518808364868,
+      "eval_overall_accuracy": 0.8359822478473619,
+      "eval_overall_f1": 0.8533090590337481,
+      "eval_overall_precision": 0.8284507872082747,
+      "eval_overall_recall": 0.8797052587675089,
+      "eval_runtime": 7.4797,
+      "eval_samples_per_second": 1137.217,
+      "eval_steps_per_second": 35.563,
+      "step": 42000
+    },
+    {
+      "epoch": 8.41,
+      "learning_rate": 7.929122945951298e-06,
+      "loss": 0.4042,
+      "step": 42500
+    },
+    {
+      "epoch": 8.51,
+      "learning_rate": 7.434171451197783e-06,
+      "loss": 0.4042,
+      "step": 43000
+    },
+    {
+      "epoch": 8.51,
+      "eval_COMMENT": {
+        "f1": 0.7073401950162513,
+        "number": 6901,
+        "precision": 0.663912546078556,
+        "recall": 0.7568468337922041
+      },
+      "eval_NAME": {
+        "f1": 0.8030025020850708,
+        "number": 8857,
+        "precision": 0.7910823838737949,
+        "recall": 0.8152873433442475
+      },
+      "eval_QTY": {
+        "f1": 0.9858200179843675,
+        "number": 7189,
+        "precision": 0.980462300495322,
+        "recall": 0.991236611489776
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7909090909090909,
+        "number": 92,
+        "precision": 0.6796875,
+        "recall": 0.9456521739130435
+      },
+      "eval_UNIT": {
+        "f1": 0.9534962089300758,
+        "number": 5732,
+        "precision": 0.921961550993809,
+        "recall": 0.9872644801116539
+      },
+      "eval_loss": 0.48657840490341187,
+      "eval_overall_accuracy": 0.8364563972233813,
+      "eval_overall_f1": 0.8538279932546375,
+      "eval_overall_precision": 0.8292443250679682,
+      "eval_overall_recall": 0.8799138020923847,
+      "eval_runtime": 9.0804,
+      "eval_samples_per_second": 936.741,
+      "eval_steps_per_second": 29.294,
+      "step": 43000
+    },
+    {
+      "epoch": 8.61,
+      "learning_rate": 6.939219956444269e-06,
+      "loss": 0.4036,
+      "step": 43500
+    },
+    {
+      "epoch": 8.71,
+      "learning_rate": 6.444268461690754e-06,
+      "loss": 0.4135,
+      "step": 44000
+    },
+    {
+      "epoch": 8.71,
+      "eval_COMMENT": {
+        "f1": 0.7091805715057758,
+        "number": 6901,
+        "precision": 0.664262212098203,
+        "recall": 0.7606144037096073
+      },
+      "eval_NAME": {
+        "f1": 0.802937576499388,
+        "number": 8857,
+        "precision": 0.7914884282110344,
+        "recall": 0.8147228181099695
+      },
+      "eval_QTY": {
+        "f1": 0.985878443859892,
+        "number": 7189,
+        "precision": 0.9812594736116853,
+        "recall": 0.9905411044651551
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7813953488372093,
+        "number": 92,
+        "precision": 0.6829268292682927,
+        "recall": 0.9130434782608695
+      },
+      "eval_UNIT": {
+        "f1": 0.9540588384051252,
+        "number": 5732,
+        "precision": 0.9230141901810471,
+        "recall": 0.9872644801116539
+      },
+      "eval_loss": 0.48304322361946106,
+      "eval_overall_accuracy": 0.8367977847741153,
+      "eval_overall_f1": 0.8542520362219861,
+      "eval_overall_precision": 0.8296429741238126,
+      "eval_overall_recall": 0.8803656459629488,
+      "eval_runtime": 6.5688,
+      "eval_samples_per_second": 1294.905,
+      "eval_steps_per_second": 40.494,
+      "step": 44000
+    },
+    {
+      "epoch": 8.81,
+      "learning_rate": 5.949316966937241e-06,
+      "loss": 0.4078,
+      "step": 44500
+    },
+    {
+      "epoch": 8.91,
+      "learning_rate": 5.4543654721837265e-06,
+      "loss": 0.3958,
+      "step": 45000
+    },
+    {
+      "epoch": 8.91,
+      "eval_COMMENT": {
+        "f1": 0.7085999057175567,
+        "number": 6901,
+        "precision": 0.6619275289380976,
+        "recall": 0.7623532821330242
+      },
+      "eval_NAME": {
+        "f1": 0.8035773802910788,
+        "number": 8857,
+        "precision": 0.7909240021869874,
+        "recall": 0.8166422039065147
+      },
+      "eval_QTY": {
+        "f1": 0.9864836764400082,
+        "number": 7189,
+        "precision": 0.9831445150594087,
+        "recall": 0.9898455974405341
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7906976744186046,
+        "number": 92,
+        "precision": 0.6910569105691057,
+        "recall": 0.9239130434782609
+      },
+      "eval_UNIT": {
+        "f1": 0.9539395984477813,
+        "number": 5732,
+        "precision": 0.9235543939888925,
+        "recall": 0.9863921842288904
+      },
+      "eval_loss": 0.4859357178211212,
+      "eval_overall_accuracy": 0.8370822743997269,
+      "eval_overall_f1": 0.8542639055049118,
+      "eval_overall_precision": 0.829048927263213,
+      "eval_overall_recall": 0.8810607903792013,
+      "eval_runtime": 9.1351,
+      "eval_samples_per_second": 931.131,
+      "eval_steps_per_second": 29.118,
+      "step": 45000
+    },
+    {
+      "epoch": 9.01,
+      "learning_rate": 4.959413977430212e-06,
+      "loss": 0.4071,
+      "step": 45500
+    },
+    {
+      "epoch": 9.11,
+      "learning_rate": 4.4644624826766974e-06,
+      "loss": 0.3985,
+      "step": 46000
+    },
+    {
+      "epoch": 9.11,
+      "eval_COMMENT": {
+        "f1": 0.7050125483280201,
+        "number": 6901,
+        "precision": 0.6627135934710533,
+        "recall": 0.7530792638748007
+      },
+      "eval_NAME": {
+        "f1": 0.8021064301552105,
+        "number": 8857,
+        "precision": 0.7878688881629097,
+        "recall": 0.8168680140002258
+      },
+      "eval_QTY": {
+        "f1": 0.9866093110386457,
+        "number": 7189,
+        "precision": 0.9842192691029901,
+        "recall": 0.989010989010989
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7853881278538813,
+        "number": 92,
+        "precision": 0.6771653543307087,
+        "recall": 0.9347826086956522
+      },
+      "eval_UNIT": {
+        "f1": 0.9550875412331895,
+        "number": 5732,
+        "precision": 0.9269413889344935,
+        "recall": 0.984996510816469
+      },
+      "eval_loss": 0.4896639585494995,
+      "eval_overall_accuracy": 0.8351856768956492,
+      "eval_overall_f1": 0.8533036226746344,
+      "eval_overall_precision": 0.8295532871631601,
+      "eval_overall_recall": 0.8784539988182545,
+      "eval_runtime": 6.8177,
+      "eval_samples_per_second": 1247.642,
+      "eval_steps_per_second": 39.016,
+      "step": 46000
+    },
+    {
+      "epoch": 9.21,
+      "learning_rate": 3.969510987923184e-06,
+      "loss": 0.3989,
+      "step": 46500
+    },
+    {
+      "epoch": 9.31,
+      "learning_rate": 3.4745594931696697e-06,
+      "loss": 0.3944,
+      "step": 47000
+    },
+    {
+      "epoch": 9.31,
+      "eval_COMMENT": {
+        "f1": 0.709349593495935,
+        "number": 6901,
+        "precision": 0.6661152818424736,
+        "recall": 0.7585857122156209
+      },
+      "eval_NAME": {
+        "f1": 0.8033533200088829,
+        "number": 8857,
+        "precision": 0.7902785363189514,
+        "recall": 0.8168680140002258
+      },
+      "eval_QTY": {
+        "f1": 0.9866833125260092,
+        "number": 7189,
+        "precision": 0.9839535205422604,
+        "recall": 0.9894282932257615
+      },
+      "eval_RANGE_END": {
+        "f1": 0.787037037037037,
+        "number": 92,
+        "precision": 0.6854838709677419,
+        "recall": 0.9239130434782609
+      },
+      "eval_UNIT": {
+        "f1": 0.954649100582721,
+        "number": 5732,
+        "precision": 0.9251923391717138,
+        "recall": 0.9860432658757851
+      },
+      "eval_loss": 0.48901429772377014,
+      "eval_overall_accuracy": 0.8369495125744415,
+      "eval_overall_f1": 0.8547268216112208,
+      "eval_overall_precision": 0.8308176926105788,
+      "eval_overall_recall": 0.8800528309756351,
+      "eval_runtime": 9.5285,
+      "eval_samples_per_second": 892.692,
+      "eval_steps_per_second": 27.916,
+      "step": 47000
+    },
+    {
+      "epoch": 9.4,
+      "learning_rate": 2.979607998416155e-06,
+      "loss": 0.397,
+      "step": 47500
+    },
+    {
+      "epoch": 9.5,
+      "learning_rate": 2.484656503662641e-06,
+      "loss": 0.3959,
+      "step": 48000
+    },
+    {
+      "epoch": 9.5,
+      "eval_COMMENT": {
+        "f1": 0.7106236964273699,
+        "number": 6901,
+        "precision": 0.6632755589047978,
+        "recall": 0.765251412838719
+      },
+      "eval_NAME": {
+        "f1": 0.8044214853080043,
+        "number": 8857,
+        "precision": 0.7917122239230264,
+        "recall": 0.8175454442813593
+      },
+      "eval_QTY": {
+        "f1": 0.9865483289418943,
+        "number": 7189,
+        "precision": 0.9835476289229919,
+        "recall": 0.9895673946306858
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7926267281105991,
+        "number": 92,
+        "precision": 0.688,
+        "recall": 0.9347826086956522
+      },
+      "eval_UNIT": {
+        "f1": 0.9542693216334796,
+        "number": 5732,
+        "precision": 0.9240196078431373,
+        "recall": 0.9865666434054431
+      },
+      "eval_loss": 0.4869215190410614,
+      "eval_overall_accuracy": 0.8368736486742784,
+      "eval_overall_f1": 0.8550634297555469,
+      "eval_overall_precision": 0.829693323742889,
+      "eval_overall_recall": 0.8820339925619547,
+      "eval_runtime": 6.6612,
+      "eval_samples_per_second": 1276.945,
+      "eval_steps_per_second": 39.933,
+      "step": 48000
+    },
+    {
+      "epoch": 9.6,
+      "learning_rate": 1.989705008909127e-06,
+      "loss": 0.3916,
+      "step": 48500
+    },
+    {
+      "epoch": 9.7,
+      "learning_rate": 1.4947535141556129e-06,
+      "loss": 0.3979,
+      "step": 49000
+    },
+    {
+      "epoch": 9.7,
+      "eval_COMMENT": {
+        "f1": 0.7113812901046943,
+        "number": 6901,
+        "precision": 0.666244939271255,
+        "recall": 0.7630778148094479
+      },
+      "eval_NAME": {
+        "f1": 0.8035317636605953,
+        "number": 8857,
+        "precision": 0.7906239755218009,
+        "recall": 0.8168680140002258
+      },
+      "eval_QTY": {
+        "f1": 0.9864780528396089,
+        "number": 7189,
+        "precision": 0.9835453539823009,
+        "recall": 0.9894282932257615
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7813953488372093,
+        "number": 92,
+        "precision": 0.6829268292682927,
+        "recall": 0.9130434782608695
+      },
+      "eval_UNIT": {
+        "f1": 0.9541810817652518,
+        "number": 5732,
+        "precision": 0.9240071907174375,
+        "recall": 0.9863921842288904
+      },
+      "eval_loss": 0.48777857422828674,
+      "eval_overall_accuracy": 0.8369115806243599,
+      "eval_overall_f1": 0.8550421585160202,
+      "eval_overall_precision": 0.8304235317239346,
+      "eval_overall_recall": 0.8811650620416391,
+      "eval_runtime": 8.6569,
+      "eval_samples_per_second": 982.572,
+      "eval_steps_per_second": 30.727,
+      "step": 49000
+    },
+    {
+      "epoch": 9.8,
+      "learning_rate": 9.998020194020988e-07,
+      "loss": 0.4088,
+      "step": 49500
+    },
+    {
+      "epoch": 9.9,
+      "learning_rate": 5.048505246485845e-07,
+      "loss": 0.4095,
+      "step": 50000
+    },
+    {
+      "epoch": 9.9,
+      "eval_COMMENT": {
+        "f1": 0.7118095366561004,
+        "number": 6901,
+        "precision": 0.665783497350492,
+        "recall": 0.7646717866975801
+      },
+      "eval_NAME": {
+        "f1": 0.80368950380619,
+        "number": 8857,
+        "precision": 0.7912472647702407,
+        "recall": 0.8165292988596591
+      },
+      "eval_QTY": {
+        "f1": 0.9863412604867227,
+        "number": 7189,
+        "precision": 0.9832734310201825,
+        "recall": 0.9894282932257615
+      },
+      "eval_RANGE_END": {
+        "f1": 0.7813953488372093,
+        "number": 92,
+        "precision": 0.6829268292682927,
+        "recall": 0.9130434782608695
+      },
+      "eval_UNIT": {
+        "f1": 0.9544226873733964,
+        "number": 5732,
+        "precision": 0.9244604316546763,
+        "recall": 0.9863921842288904
+      },
+      "eval_loss": 0.48620378971099854,
+      "eval_overall_accuracy": 0.8372719341501347,
+      "eval_overall_f1": 0.855167762603271,
+      "eval_overall_precision": 0.8304135695340384,
+      "eval_overall_recall": 0.8814431198081402,
+      "eval_runtime": 8.2871,
+      "eval_samples_per_second": 1026.42,
+      "eval_steps_per_second": 32.098,
+      "step": 50000
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 9.899029895070283e-09,
+      "loss": 0.3994,
+      "step": 50500
+    },
+    {
+      "epoch": 10.0,
+      "step": 50510,
+      "total_flos": 1555191327435270.0,
+      "train_loss": 0.45639907907811844,
+      "train_runtime": 1954.9419,
+      "train_samples_per_second": 826.638,
+      "train_steps_per_second": 25.837
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 50510,
+  "num_train_epochs": 10,
+  "save_steps": 1000,
+  "total_flos": 1555191327435270.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8c5410c47250eb6dc78a5b7266e040300f9e0d9b784546b7746f4dfdda92fce2
+size 4600

validation_results.json

@@ -0,0 +1,41 @@
+{
+    "epoch": 10.0,
+    "eval_COMMENT": {
+        "f1": 0.69833923223523,
+        "number": 6901,
+        "precision": 0.6584520600693108,
+        "recall": 0.7433705260107231
+    },
+    "eval_NAME": {
+        "f1": 0.8048291977300545,
+        "number": 8857,
+        "precision": 0.7933530766699572,
+        "recall": 0.8166422039065147
+    },
+    "eval_QTY": {
+        "f1": 0.9856135011758197,
+        "number": 7189,
+        "precision": 0.9801898472967395,
+        "recall": 0.9910975100848518
+    },
+    "eval_RANGE_END": {
+        "f1": 0.7837837837837838,
+        "number": 92,
+        "precision": 0.6692307692307692,
+        "recall": 0.9456521739130435
+    },
+    "eval_UNIT": {
+        "f1": 0.9537942664418213,
+        "number": 5732,
+        "precision": 0.922976501305483,
+        "recall": 0.9867411025819958
+    },
+    "eval_loss": 0.47403302788734436,
+    "eval_overall_accuracy": 0.8347115275196297,
+    "eval_overall_f1": 0.8523122656487518,
+    "eval_overall_precision": 0.8290126499096435,
+    "eval_overall_recall": 0.8769594383233117,
+    "eval_runtime": 9.862,
+    "eval_samples_per_second": 862.507,
+    "eval_steps_per_second": 26.972
+}