End of training

README.md

@@ -16,15 +16,16 @@ should probably proofread and complete it, then remove this comment. -->
 
 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: 2.5417
-- Comment: {'precision': 0.5882352941176471, 'recall': 0.8333333333333334, 'f1': 0.6896551724137931, 'number': 12}
-- Name: {'precision': 0.7692307692307693, 'recall': 0.7692307692307693, 'f1': 0.7692307692307693, 'number': 26}
-- Qty: {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 22}
-- Unit: {'precision': 0.8947368421052632, 'recall': 0.9444444444444444, 'f1': 0.918918918918919, 'number': 18}
-- Overall Precision: 0.8214
-- Overall Recall: 0.8846
-- Overall F1: 0.8519
-- Overall Accuracy: 0.7619
+- Loss: 2.5723
+- Comment: {'precision': 0.6705310396409873, 'recall': 0.7578191039729502, 'f1': 0.7115079365079365, 'number': 7098}
+- Name: {'precision': 0.8150406504065041, 'recall': 0.8209244693459756, 'f1': 0.8179719791722584, 'number': 9281}
+- Qty: {'precision': 0.9861000794281175, 'recall': 0.9857086145295753, 'f1': 0.9859043081199126, 'number': 7557}
+- Range End: {'precision': 0.5986842105263158, 'recall': 0.9479166666666666, 'f1': 0.7338709677419355, 'number': 96}
+- Unit: {'precision': 0.9225395839801304, 'recall': 0.985735611212473, 'f1': 0.9530911715179216, 'number': 6029}
+- Overall Precision: 0.8402
+- Overall Recall: 0.8809
+- Overall F1: 0.8601
+- Overall Accuracy: 0.8330
 
 ## Model description
 
@@ -53,23 +54,23 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step  | Validation Loss | Comment                                                                                                  | Name                                                                                                    | Qty                                                        | Range End                                                                | Unit                                                                                     | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
-|:-------------:|:-----:|:-----:|:---------------:|:--------------------------------------------------------------------------------------------------------:|:-------------------------------------------------------------------------------------------------------:|:----------------------------------------------------------:|:------------------------------------------------------------------------:|:----------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
-| 4.2414        | 0.2   | 1000  | 3.7464          | {'precision': 0.45454545454545453, 'recall': 0.4411764705882353, 'f1': 0.4477611940298507, 'number': 34} | {'precision': 0.5806451612903226, 'recall': 0.6923076923076923, 'f1': 0.631578947368421, 'number': 26}  | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 3}                | {'precision': 1.0, 'recall': 0.9047619047619048, 'f1': 0.9500000000000001, 'number': 21} | 0.7075            | 0.7009         | 0.7042     | 0.7353           |
-| 3.5305        | 0.4   | 2000  | 3.3135          | {'precision': 0.5161290322580645, 'recall': 0.47058823529411764, 'f1': 0.4923076923076923, 'number': 34} | {'precision': 0.5806451612903226, 'recall': 0.6923076923076923, 'f1': 0.631578947368421, 'number': 26}  | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 1.0, 'recall': 0.3333333333333333, 'f1': 0.5, 'number': 3} | {'precision': 1.0, 'recall': 0.9047619047619048, 'f1': 0.9500000000000001, 'number': 21} | 0.7333            | 0.7196         | 0.7264     | 0.7892           |
-| 3.2255        | 0.59  | 3000  | 3.0933          | {'precision': 0.5555555555555556, 'recall': 0.4411764705882353, 'f1': 0.49180327868852464, 'number': 34} | {'precision': 0.6428571428571429, 'recall': 0.6923076923076923, 'f1': 0.6666666666666666, 'number': 26} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 3}                | {'precision': 1.0, 'recall': 0.9047619047619048, 'f1': 0.9500000000000001, 'number': 21} | 0.78              | 0.7290         | 0.7536     | 0.7598           |
-| 3.073         | 0.79  | 4000  | 2.9998          | {'precision': 0.5, 'recall': 0.4117647058823529, 'f1': 0.45161290322580644, 'number': 34}                | {'precision': 0.6206896551724138, 'recall': 0.6923076923076923, 'f1': 0.6545454545454545, 'number': 26} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 3}                | {'precision': 1.0, 'recall': 0.9047619047619048, 'f1': 0.9500000000000001, 'number': 21} | 0.7549            | 0.7196         | 0.7368     | 0.7402           |
-| 2.954         | 0.99  | 5000  | 2.8983          | {'precision': 0.7307692307692307, 'recall': 0.5588235294117647, 'f1': 0.6333333333333334, 'number': 34}  | {'precision': 0.6785714285714286, 'recall': 0.7307692307692307, 'f1': 0.7037037037037038, 'number': 26} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 1.0, 'recall': 0.3333333333333333, 'f1': 0.5, 'number': 3} | {'precision': 1.0, 'recall': 0.9523809523809523, 'f1': 0.975609756097561, 'number': 21}  | 0.8367            | 0.7664         | 0.8000     | 0.7941           |
-| 2.8463        | 1.19  | 6000  | 2.8088          | {'precision': 0.5714285714285714, 'recall': 0.47058823529411764, 'f1': 0.5161290322580646, 'number': 34} | {'precision': 0.6428571428571429, 'recall': 0.6923076923076923, 'f1': 0.6666666666666666, 'number': 26} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 3}                | {'precision': 1.0, 'recall': 0.9047619047619048, 'f1': 0.9500000000000001, 'number': 21} | 0.7822            | 0.7383         | 0.7596     | 0.7696           |
-| 2.7953        | 1.39  | 7000  | 2.7729          | {'precision': 0.6153846153846154, 'recall': 0.47058823529411764, 'f1': 0.5333333333333333, 'number': 34} | {'precision': 0.6666666666666666, 'recall': 0.6923076923076923, 'f1': 0.6792452830188679, 'number': 26} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 3}                | {'precision': 1.0, 'recall': 0.9047619047619048, 'f1': 0.9500000000000001, 'number': 21} | 0.8061            | 0.7383         | 0.7707     | 0.7647           |
-| 2.7561        | 1.58  | 8000  | 2.7151          | {'precision': 0.6666666666666666, 'recall': 0.5294117647058824, 'f1': 0.5901639344262295, 'number': 34}  | {'precision': 0.6206896551724138, 'recall': 0.6923076923076923, 'f1': 0.6545454545454545, 'number': 26} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 1.0, 'recall': 0.6666666666666666, 'f1': 0.8, 'number': 3} | {'precision': 1.0, 'recall': 0.9523809523809523, 'f1': 0.975609756097561, 'number': 21}  | 0.8020            | 0.7570         | 0.7788     | 0.7696           |
-| 2.789         | 1.78  | 9000  | 2.6807          | {'precision': 0.6538461538461539, 'recall': 0.5, 'f1': 0.5666666666666668, 'number': 34}                 | {'precision': 0.5666666666666667, 'recall': 0.6538461538461539, 'f1': 0.6071428571428571, 'number': 26} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 3}                | {'precision': 1.0, 'recall': 0.9047619047619048, 'f1': 0.9500000000000001, 'number': 21} | 0.7822            | 0.7383         | 0.7596     | 0.7598           |
-| 2.6709        | 1.98  | 10000 | 2.6451          | {'precision': 0.6153846153846154, 'recall': 0.47058823529411764, 'f1': 0.5333333333333333, 'number': 34} | {'precision': 0.6, 'recall': 0.6923076923076923, 'f1': 0.6428571428571429, 'number': 26}                | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 3}                | {'precision': 1.0, 'recall': 0.9047619047619048, 'f1': 0.9500000000000001, 'number': 21} | 0.7822            | 0.7383         | 0.7596     | 0.7696           |
-| 2.5318        | 2.18  | 11000 | 2.6231          | {'precision': 0.5555555555555556, 'recall': 0.4411764705882353, 'f1': 0.49180327868852464, 'number': 34} | {'precision': 0.6333333333333333, 'recall': 0.7307692307692307, 'f1': 0.6785714285714285, 'number': 26} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 3}                | {'precision': 1.0, 'recall': 0.9523809523809523, 'f1': 0.975609756097561, 'number': 21}  | 0.7767            | 0.7477         | 0.7619     | 0.7745           |
-| 2.6576        | 2.38  | 12000 | 2.6023          | {'precision': 0.6923076923076923, 'recall': 0.5294117647058824, 'f1': 0.5999999999999999, 'number': 34}  | {'precision': 0.7037037037037037, 'recall': 0.7307692307692307, 'f1': 0.7169811320754716, 'number': 26} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 3}                | {'precision': 1.0, 'recall': 0.9523809523809523, 'f1': 0.975609756097561, 'number': 21}  | 0.8384            | 0.7757         | 0.8058     | 0.7843           |
-| 2.5836        | 2.57  | 13000 | 2.5741          | {'precision': 0.68, 'recall': 0.5, 'f1': 0.576271186440678, 'number': 34}                                | {'precision': 0.6551724137931034, 'recall': 0.7307692307692307, 'f1': 0.6909090909090909, 'number': 26} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 3}                | {'precision': 1.0, 'recall': 0.9523809523809523, 'f1': 0.975609756097561, 'number': 21}  | 0.82              | 0.7664         | 0.7923     | 0.7892           |
-| 2.6004        | 2.77  | 14000 | 2.5654          | {'precision': 0.6538461538461539, 'recall': 0.5, 'f1': 0.5666666666666668, 'number': 34}                 | {'precision': 0.6, 'recall': 0.6923076923076923, 'f1': 0.6428571428571429, 'number': 26}                | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 3}                | {'precision': 1.0, 'recall': 0.9047619047619048, 'f1': 0.9500000000000001, 'number': 21} | 0.7921            | 0.7477         | 0.7692     | 0.7745           |
-| 2.5449        | 2.97  | 15000 | 2.5586          | {'precision': 0.72, 'recall': 0.5294117647058824, 'f1': 0.6101694915254237, 'number': 34}                | {'precision': 0.6206896551724138, 'recall': 0.6923076923076923, 'f1': 0.6545454545454545, 'number': 26} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 23} | {'precision': 1.0, 'recall': 1.0, 'f1': 1.0, 'number': 3}                | {'precision': 1.0, 'recall': 0.9047619047619048, 'f1': 0.9500000000000001, 'number': 21} | 0.8182            | 0.7570         | 0.7864     | 0.7892           |
+| Training Loss | Epoch | Step  | Validation Loss | Comment                                                                                                   | Name                                                                                                      | Qty                                                                                                       | Range End                                                                                                 | Unit                                                                                                      | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:---------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------------------:|:-----------------:|:--------------:|:----------:|:----------------:|
+| 4.174         | 0.2   | 1000  | 3.8690          | {'precision': 0.5304157015725954, 'recall': 0.6285755561976307, 'f1': 0.5753388429752067, 'number': 6922} | {'precision': 0.7673592421143288, 'recall': 0.8069738480697385, 'f1': 0.7866681381745945, 'number': 8833} | {'precision': 0.9667778704475952, 'recall': 0.9806824591088551, 'f1': 0.9736805263894722, 'number': 7092} | {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'number': 88}                                                | {'precision': 0.9121887287024901, 'recall': 0.9756439460311898, 'f1': 0.9428498856997714, 'number': 5707} | 0.7795            | 0.8380         | 0.8077     | 0.7962           |
+| 3.5528        | 0.4   | 2000  | 3.4154          | {'precision': 0.5496301598663803, 'recall': 0.6655590869690841, 'f1': 0.6020648196549921, 'number': 6922} | {'precision': 0.7787928221859707, 'recall': 0.8107098381070984, 'f1': 0.7944308852895496, 'number': 8833} | {'precision': 0.9785673998871969, 'recall': 0.9785673998871969, 'f1': 0.9785673998871969, 'number': 7092} | {'precision': 0.6666666666666666, 'recall': 0.13636363636363635, 'f1': 0.22641509433962262, 'number': 88} | {'precision': 0.9109700520833334, 'recall': 0.9807254249167688, 'f1': 0.9445616403679015, 'number': 5707} | 0.7887            | 0.8490         | 0.8177     | 0.8042           |
+| 3.3333        | 0.59  | 3000  | 3.1915          | {'precision': 0.5850767928407304, 'recall': 0.6989309448136377, 'f1': 0.6369560924231453, 'number': 6922} | {'precision': 0.7759810263044415, 'recall': 0.814898675421714, 'f1': 0.7949638301397095, 'number': 8833}  | {'precision': 0.9797125950972105, 'recall': 0.9805414551607445, 'f1': 0.9801268498942918, 'number': 7092} | {'precision': 0.6867469879518072, 'recall': 0.6477272727272727, 'f1': 0.6666666666666667, 'number': 88}   | {'precision': 0.9255372313843079, 'recall': 0.9735412651130191, 'f1': 0.948932536293766, 'number': 5707}  | 0.8006            | 0.8590         | 0.8288     | 0.8124           |
+| 3.1122        | 0.79  | 4000  | 3.0560          | {'precision': 0.6020151133501259, 'recall': 0.7250794568043918, 'f1': 0.6578412740022282, 'number': 6922} | {'precision': 0.7925502692011867, 'recall': 0.8165968527114231, 'f1': 0.8043938887030222, 'number': 8833} | {'precision': 0.9816358242689646, 'recall': 0.9798364354201917, 'f1': 0.9807353044950956, 'number': 7092} | {'precision': 0.5547445255474452, 'recall': 0.8636363636363636, 'f1': 0.6755555555555556, 'number': 88}   | {'precision': 0.9180758496141849, 'recall': 0.9798493078675311, 'f1': 0.9479572808950669, 'number': 5707} | 0.8082            | 0.8676         | 0.8368     | 0.8183           |
+| 3.074         | 0.99  | 5000  | 2.9495          | {'precision': 0.6171328671328671, 'recall': 0.7139555041895406, 'f1': 0.6620227729403884, 'number': 6922} | {'precision': 0.8028043623414199, 'recall': 0.816710064530737, 'f1': 0.8096975138896684, 'number': 8833}  | {'precision': 0.9747242005306521, 'recall': 0.9842075578116187, 'f1': 0.9794429242966393, 'number': 7092} | {'precision': 0.6448598130841121, 'recall': 0.7840909090909091, 'f1': 0.7076923076923077, 'number': 88}   | {'precision': 0.9168310322156475, 'recall': 0.9773961801296653, 'f1': 0.9461453651089815, 'number': 5707} | 0.8167            | 0.8653         | 0.8403     | 0.8209           |
+| 2.936         | 1.19  | 6000  | 2.8893          | {'precision': 0.6245376865195765, 'recall': 0.7074544929211211, 'f1': 0.6634152949942423, 'number': 6922} | {'precision': 0.8003099402258136, 'recall': 0.81852145363976, 'f1': 0.8093132590809874, 'number': 8833}   | {'precision': 0.9721951897678298, 'recall': 0.9860406091370558, 'f1': 0.9790689534476724, 'number': 7092} | {'precision': 0.6220472440944882, 'recall': 0.8977272727272727, 'f1': 0.7348837209302326, 'number': 88}   | {'precision': 0.9140714169248328, 'recall': 0.9823024356053969, 'f1': 0.9469594594594595, 'number': 5707} | 0.8179            | 0.8660         | 0.8413     | 0.8217           |
+| 2.7662        | 1.39  | 7000  | 2.8622          | {'precision': 0.6298537569339385, 'recall': 0.7217567177116441, 'f1': 0.6726807593914097, 'number': 6922} | {'precision': 0.7999777753083676, 'recall': 0.8150118872410279, 'f1': 0.8074248541947062, 'number': 8833} | {'precision': 0.9800337457817773, 'recall': 0.9827975183305132, 'f1': 0.9814136862855534, 'number': 7092} | {'precision': 0.6290322580645161, 'recall': 0.8863636363636364, 'f1': 0.7358490566037735, 'number': 88}   | {'precision': 0.9191902567478605, 'recall': 0.9786227439985982, 'f1': 0.9479758974794195, 'number': 5707} | 0.8210            | 0.8668         | 0.8433     | 0.8235           |
+| 2.7839        | 1.58  | 8000  | 2.7801          | {'precision': 0.6325475860330266, 'recall': 0.7249349898873158, 'f1': 0.6755974419387412, 'number': 6922} | {'precision': 0.8036190053285968, 'recall': 0.8195403600135854, 'f1': 0.8115015974440895, 'number': 8833} | {'precision': 0.975977653631285, 'recall': 0.9853355893965031, 'f1': 0.9806342969407802, 'number': 7092}  | {'precision': 0.6521739130434783, 'recall': 0.8522727272727273, 'f1': 0.7389162561576356, 'number': 88}   | {'precision': 0.9188301018731515, 'recall': 0.9798493078675311, 'f1': 0.9483591961332994, 'number': 5707} | 0.8221            | 0.8698         | 0.8453     | 0.8242           |
+| 2.7221        | 1.78  | 9000  | 2.7520          | {'precision': 0.6436781609195402, 'recall': 0.7442935567754985, 'f1': 0.690339005761758, 'number': 6922}  | {'precision': 0.8124719605204127, 'recall': 0.8201064191101551, 'f1': 0.8162713392303792, 'number': 8833} | {'precision': 0.9827975183305132, 'recall': 0.9827975183305132, 'f1': 0.9827975183305132, 'number': 7092} | {'precision': 0.6576576576576577, 'recall': 0.8295454545454546, 'f1': 0.7336683417085428, 'number': 88}   | {'precision': 0.9227716222920457, 'recall': 0.9777466269493604, 'f1': 0.9494640122511486, 'number': 5707} | 0.8293            | 0.8735         | 0.8508     | 0.8285           |
+| 2.7156        | 1.98  | 10000 | 2.7236          | {'precision': 0.6453828542355635, 'recall': 0.7330251372435712, 'f1': 0.6864177489177489, 'number': 6922} | {'precision': 0.8084821428571428, 'recall': 0.8201064191101551, 'f1': 0.8142527960433878, 'number': 8833} | {'precision': 0.9825204398082887, 'recall': 0.9827975183305132, 'f1': 0.9826589595375722, 'number': 7092} | {'precision': 0.6324786324786325, 'recall': 0.8409090909090909, 'f1': 0.7219512195121951, 'number': 88}   | {'precision': 0.9222387320455672, 'recall': 0.9787979674084458, 'f1': 0.9496769806188372, 'number': 5707} | 0.8291            | 0.8710         | 0.8496     | 0.8265           |
+| 2.6804        | 2.18  | 11000 | 2.6929          | {'precision': 0.6422784494578088, 'recall': 0.7444380236925744, 'f1': 0.6895951823352291, 'number': 6922} | {'precision': 0.8120670391061453, 'recall': 0.8228235027736895, 'f1': 0.8174098858460327, 'number': 8833} | {'precision': 0.9837570621468926, 'recall': 0.9820924985899605, 'f1': 0.9829240756421113, 'number': 7092} | {'precision': 0.635593220338983, 'recall': 0.8522727272727273, 'f1': 0.7281553398058253, 'number': 88}    | {'precision': 0.9234584228798148, 'recall': 0.9787979674084458, 'f1': 0.9503232391970058, 'number': 5707} | 0.8288            | 0.8745         | 0.8510     | 0.8279           |
+| 2.6121        | 2.38  | 12000 | 2.6691          | {'precision': 0.6490939044481054, 'recall': 0.7399595492632187, 'f1': 0.691554715452643, 'number': 6922}  | {'precision': 0.811037725288257, 'recall': 0.820219630929469, 'f1': 0.8156028368794326, 'number': 8833}   | {'precision': 0.9798121407542408, 'recall': 0.9854765933446137, 'f1': 0.9826362038664324, 'number': 7092} | {'precision': 0.6328125, 'recall': 0.9204545454545454, 'f1': 0.7499999999999999, 'number': 88}            | {'precision': 0.9207106431978944, 'recall': 0.9807254249167688, 'f1': 0.9497709146444936, 'number': 5707} | 0.8299            | 0.8740         | 0.8514     | 0.8277           |
+| 2.553         | 2.57  | 13000 | 2.6652          | {'precision': 0.6478233438485804, 'recall': 0.7416931522681306, 'f1': 0.6915875260995488, 'number': 6922} | {'precision': 0.8128150554497592, 'recall': 0.8214649609419223, 'f1': 0.8171171171171171, 'number': 8833} | {'precision': 0.9826760563380281, 'recall': 0.983784545967287, 'f1': 0.9832299887260428, 'number': 7092}  | {'precision': 0.639344262295082, 'recall': 0.8863636363636364, 'f1': 0.742857142857143, 'number': 88}     | {'precision': 0.920335085413929, 'recall': 0.9817767653758542, 'f1': 0.9500635862653668, 'number': 5707}  | 0.8304            | 0.8745         | 0.8519     | 0.8287           |
+| 2.5781        | 2.77  | 14000 | 2.6431          | {'precision': 0.6512514292974209, 'recall': 0.7405374169315226, 'f1': 0.6930304873926858, 'number': 6922} | {'precision': 0.8114304887596465, 'recall': 0.8213517491226084, 'f1': 0.8163609767075503, 'number': 8833} | {'precision': 0.9832252607837609, 'recall': 0.983502538071066, 'f1': 0.9833638798815735, 'number': 7092}  | {'precision': 0.6551724137931034, 'recall': 0.8636363636363636, 'f1': 0.7450980392156864, 'number': 88}   | {'precision': 0.9220500988793672, 'recall': 0.9803749780970737, 'f1': 0.9503184713375797, 'number': 5707} | 0.8317            | 0.8738         | 0.8522     | 0.8286           |
+| 2.5928        | 2.97  | 15000 | 2.6394          | {'precision': 0.6551064643631264, 'recall': 0.7422710199364345, 'f1': 0.6959701997968167, 'number': 6922} | {'precision': 0.8126049479458188, 'recall': 0.8218045963998641, 'f1': 0.8171788810086682, 'number': 8833} | {'precision': 0.9825376707505985, 'recall': 0.983784545967287, 'f1': 0.9831607130275488, 'number': 7092}  | {'precision': 0.6470588235294118, 'recall': 0.875, 'f1': 0.7439613526570048, 'number': 88}                | {'precision': 0.9219729462223688, 'recall': 0.9793236376379885, 'f1': 0.9497833290848839, 'number': 5707} | 0.8331            | 0.8742         | 0.8532     | 0.8289           |
 
 
 ### Framework versions

all_results.json

@@ -1,41 +1,41 @@
 {
     "epoch": 3.0,
     "eval_COMMENT": {
-        "f1": 0.6896551724137931,
-        "number": 12,
-        "precision": 0.5882352941176471,
-        "recall": 0.8333333333333334
+        "f1": 0.7115079365079365,
+        "number": 7098,
+        "precision": 0.6705310396409873,
+        "recall": 0.7578191039729502
     },
     "eval_NAME": {
-        "f1": 0.7692307692307693,
-        "number": 26,
-        "precision": 0.7692307692307693,
-        "recall": 0.7692307692307693
+        "f1": 0.8179719791722584,
+        "number": 9281,
+        "precision": 0.8150406504065041,
+        "recall": 0.8209244693459756
     },
     "eval_QTY": {
-        "f1": 1.0,
-        "number": 22,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9859043081199126,
+        "number": 7557,
+        "precision": 0.9861000794281175,
+        "recall": 0.9857086145295753
     },
     "eval_RANGE_END": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.7338709677419355,
+        "number": 96,
+        "precision": 0.5986842105263158,
+        "recall": 0.9479166666666666
     },
     "eval_UNIT": {
-        "f1": 0.918918918918919,
-        "number": 18,
-        "precision": 0.8947368421052632,
-        "recall": 0.9444444444444444
+        "f1": 0.9530911715179216,
+        "number": 6029,
+        "precision": 0.9225395839801304,
+        "recall": 0.985735611212473
     },
-    "eval_loss": 2.54168963432312,
-    "eval_overall_accuracy": 0.7619047619047619,
-    "eval_overall_f1": 0.8518518518518519,
-    "eval_overall_precision": 0.8214285714285714,
-    "eval_overall_recall": 0.8846153846153846,
-    "eval_runtime": 11.2254,
-    "eval_samples_per_second": 797.652,
-    "eval_steps_per_second": 24.943
+    "eval_loss": 2.572307586669922,
+    "eval_overall_accuracy": 0.8329675956404067,
+    "eval_overall_f1": 0.8600659315675797,
+    "eval_overall_precision": 0.8401865600609175,
+    "eval_overall_recall": 0.8809088187352383,
+    "eval_runtime": 10.0829,
+    "eval_samples_per_second": 888.036,
+    "eval_steps_per_second": 27.77
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:75a1845aa8742790e488f470021895268092cc242fa2e9ad4f5932ef21215ba0
-size 69005023
+oid sha256:4fe51036b0fe145c1633527037e9fd147f46ac6333b510111d93b14646958829
+size 69005087

test_results.json

@@ -1,35 +1,41 @@
 {
     "epoch": 3.0,
     "eval_COMMENT": {
-        "f1": 0.6896551724137931,
-        "number": 12,
-        "precision": 0.5882352941176471,
-        "recall": 0.8333333333333334
+        "f1": 0.7115079365079365,
+        "number": 7098,
+        "precision": 0.6705310396409873,
+        "recall": 0.7578191039729502
     },
     "eval_NAME": {
-        "f1": 0.7692307692307693,
-        "number": 26,
-        "precision": 0.7692307692307693,
-        "recall": 0.7692307692307693
+        "f1": 0.8179719791722584,
+        "number": 9281,
+        "precision": 0.8150406504065041,
+        "recall": 0.8209244693459756
     },
     "eval_QTY": {
-        "f1": 1.0,
-        "number": 22,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9859043081199126,
+        "number": 7557,
+        "precision": 0.9861000794281175,
+        "recall": 0.9857086145295753
+    },
+    "eval_RANGE_END": {
+        "f1": 0.7338709677419355,
+        "number": 96,
+        "precision": 0.5986842105263158,
+        "recall": 0.9479166666666666
     },
     "eval_UNIT": {
-        "f1": 0.918918918918919,
-        "number": 18,
-        "precision": 0.8947368421052632,
-        "recall": 0.9444444444444444
+        "f1": 0.9530911715179216,
+        "number": 6029,
+        "precision": 0.9225395839801304,
+        "recall": 0.985735611212473
     },
-    "eval_loss": 2.54168963432312,
-    "eval_overall_accuracy": 0.7619047619047619,
-    "eval_overall_f1": 0.8518518518518519,
-    "eval_overall_precision": 0.8214285714285714,
-    "eval_overall_recall": 0.8846153846153846,
-    "eval_runtime": 11.2254,
-    "eval_samples_per_second": 797.652,
-    "eval_steps_per_second": 24.943
+    "eval_loss": 2.572307586669922,
+    "eval_overall_accuracy": 0.8329675956404067,
+    "eval_overall_f1": 0.8600659315675797,
+    "eval_overall_precision": 0.8401865600609175,
+    "eval_overall_recall": 0.8809088187352383,
+    "eval_runtime": 10.0829,
+    "eval_samples_per_second": 888.036,
+    "eval_steps_per_second": 27.77
 }

train_results.json

@@ -1,29 +1,41 @@
 {
     "epoch": 3.0,
+    "eval_COMMENT": {
+        "f1": 0.721799355413838,
+        "number": 129768,
+        "precision": 0.6788247832786863,
+        "recall": 0.7705828863818507
+    },
     "eval_NAME": {
-        "f1": 0.8571428571428571,
-        "number": 4,
-        "precision": 1.0,
-        "recall": 0.75
+        "f1": 0.8235496724923245,
+        "number": 167686,
+        "precision": 0.8201064458900059,
+        "recall": 0.8270219338525577
     },
     "eval_QTY": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.984932056268375,
+        "number": 135815,
+        "precision": 0.9844285567177133,
+        "recall": 0.9854360711261643
+    },
+    "eval_RANGE_END": {
+        "f1": 0.7352039613298751,
+        "number": 1680,
+        "precision": 0.6087465833658727,
+        "recall": 0.9279761904761905
     },
     "eval_UNIT": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9542982760615527,
+        "number": 108817,
+        "precision": 0.9251343819294052,
+        "recall": 0.9853607432662176
     },
-    "eval_loss": 2.457996129989624,
-    "eval_overall_accuracy": 0.9629629629629629,
-    "eval_overall_f1": 0.9473684210526316,
-    "eval_overall_precision": 1.0,
-    "eval_overall_recall": 0.9,
-    "eval_runtime": 183.4069,
-    "eval_samples_per_second": 881.117,
-    "eval_steps_per_second": 27.54
+    "eval_loss": 2.4726200103759766,
+    "eval_overall_accuracy": 0.8400964270928377,
+    "eval_overall_f1": 0.8636297978361569,
+    "eval_overall_precision": 0.8431603381071257,
+    "eval_overall_recall": 0.8851178631985082,
+    "eval_runtime": 178.5371,
+    "eval_samples_per_second": 905.151,
+    "eval_steps_per_second": 28.291
 }

trainer_state.json

@@ -11,828 +11,828 @@
     {
       "epoch": 0.1,
       "learning_rate": 4.835016168415495e-05,
-      "loss": 6.8417,
+      "loss": 6.7872,
       "step": 500
     },
     {
       "epoch": 0.2,
       "learning_rate": 4.670032336830991e-05,
-      "loss": 4.2414,
+      "loss": 4.174,
       "step": 1000
     },
     {
       "epoch": 0.2,
       "eval_COMMENT": {
-        "f1": 0.4477611940298507,
-        "number": 34,
-        "precision": 0.45454545454545453,
-        "recall": 0.4411764705882353
+        "f1": 0.5753388429752067,
+        "number": 6922,
+        "precision": 0.5304157015725954,
+        "recall": 0.6285755561976307
       },
       "eval_NAME": {
-        "f1": 0.631578947368421,
-        "number": 26,
-        "precision": 0.5806451612903226,
-        "recall": 0.6923076923076923
+        "f1": 0.7866681381745945,
+        "number": 8833,
+        "precision": 0.7673592421143288,
+        "recall": 0.8069738480697385
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9736805263894722,
+        "number": 7092,
+        "precision": 0.9667778704475952,
+        "recall": 0.9806824591088551
       },
       "eval_RANGE_END": {
         "f1": 0.0,
-        "number": 3,
+        "number": 88,
         "precision": 0.0,
         "recall": 0.0
       },
       "eval_UNIT": {
-        "f1": 0.9500000000000001,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9047619047619048
-      },
-      "eval_loss": 3.7463533878326416,
-      "eval_overall_accuracy": 0.7352941176470589,
-      "eval_overall_f1": 0.704225352112676,
-      "eval_overall_precision": 0.7075471698113207,
-      "eval_overall_recall": 0.7009345794392523,
-      "eval_runtime": 9.4242,
-      "eval_samples_per_second": 902.571,
-      "eval_steps_per_second": 28.225,
+        "f1": 0.9428498856997714,
+        "number": 5707,
+        "precision": 0.9121887287024901,
+        "recall": 0.9756439460311898
+      },
+      "eval_loss": 3.8689863681793213,
+      "eval_overall_accuracy": 0.7961931378288045,
+      "eval_overall_f1": 0.8077130165567371,
+      "eval_overall_precision": 0.7795388113023709,
+      "eval_overall_recall": 0.838000139655052,
+      "eval_runtime": 10.0032,
+      "eval_samples_per_second": 850.33,
+      "eval_steps_per_second": 26.592,
       "step": 1000
     },
     {
       "epoch": 0.3,
       "learning_rate": 4.505048505246486e-05,
-      "loss": 3.6927,
+      "loss": 3.6785,
       "step": 1500
     },
     {
       "epoch": 0.4,
       "learning_rate": 4.3400646736619816e-05,
-      "loss": 3.5305,
+      "loss": 3.5528,
       "step": 2000
     },
     {
       "epoch": 0.4,
       "eval_COMMENT": {
-        "f1": 0.4923076923076923,
-        "number": 34,
-        "precision": 0.5161290322580645,
-        "recall": 0.47058823529411764
+        "f1": 0.6020648196549921,
+        "number": 6922,
+        "precision": 0.5496301598663803,
+        "recall": 0.6655590869690841
       },
       "eval_NAME": {
-        "f1": 0.631578947368421,
-        "number": 26,
-        "precision": 0.5806451612903226,
-        "recall": 0.6923076923076923
+        "f1": 0.7944308852895496,
+        "number": 8833,
+        "precision": 0.7787928221859707,
+        "recall": 0.8107098381070984
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9785673998871969,
+        "number": 7092,
+        "precision": 0.9785673998871969,
+        "recall": 0.9785673998871969
       },
       "eval_RANGE_END": {
-        "f1": 0.5,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 0.3333333333333333
+        "f1": 0.22641509433962262,
+        "number": 88,
+        "precision": 0.6666666666666666,
+        "recall": 0.13636363636363635
       },
       "eval_UNIT": {
-        "f1": 0.9500000000000001,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9047619047619048
-      },
-      "eval_loss": 3.313516139984131,
-      "eval_overall_accuracy": 0.7892156862745098,
-      "eval_overall_f1": 0.7264150943396226,
-      "eval_overall_precision": 0.7333333333333333,
-      "eval_overall_recall": 0.719626168224299,
-      "eval_runtime": 11.1909,
-      "eval_samples_per_second": 760.084,
-      "eval_steps_per_second": 23.769,
+        "f1": 0.9445616403679015,
+        "number": 5707,
+        "precision": 0.9109700520833334,
+        "recall": 0.9807254249167688
+      },
+      "eval_loss": 3.415388822555542,
+      "eval_overall_accuracy": 0.8041807504201522,
+      "eval_overall_f1": 0.8177492307433626,
+      "eval_overall_precision": 0.788719146313775,
+      "eval_overall_recall": 0.8489979750017457,
+      "eval_runtime": 10.2481,
+      "eval_samples_per_second": 830.011,
+      "eval_steps_per_second": 25.956,
       "step": 2000
     },
     {
       "epoch": 0.49,
       "learning_rate": 4.1750808420774766e-05,
-      "loss": 3.364,
+      "loss": 3.3872,
       "step": 2500
     },
     {
       "epoch": 0.59,
       "learning_rate": 4.010097010492972e-05,
-      "loss": 3.2255,
+      "loss": 3.3333,
       "step": 3000
     },
     {
       "epoch": 0.59,
       "eval_COMMENT": {
-        "f1": 0.49180327868852464,
-        "number": 34,
-        "precision": 0.5555555555555556,
-        "recall": 0.4411764705882353
+        "f1": 0.6369560924231453,
+        "number": 6922,
+        "precision": 0.5850767928407304,
+        "recall": 0.6989309448136377
       },
       "eval_NAME": {
-        "f1": 0.6666666666666666,
-        "number": 26,
-        "precision": 0.6428571428571429,
-        "recall": 0.6923076923076923
+        "f1": 0.7949638301397095,
+        "number": 8833,
+        "precision": 0.7759810263044415,
+        "recall": 0.814898675421714
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9801268498942918,
+        "number": 7092,
+        "precision": 0.9797125950972105,
+        "recall": 0.9805414551607445
       },
       "eval_RANGE_END": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.6666666666666667,
+        "number": 88,
+        "precision": 0.6867469879518072,
+        "recall": 0.6477272727272727
       },
       "eval_UNIT": {
-        "f1": 0.9500000000000001,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9047619047619048
-      },
-      "eval_loss": 3.0932886600494385,
-      "eval_overall_accuracy": 0.7598039215686274,
-      "eval_overall_f1": 0.7536231884057971,
-      "eval_overall_precision": 0.78,
-      "eval_overall_recall": 0.7289719626168224,
-      "eval_runtime": 9.5483,
-      "eval_samples_per_second": 890.84,
-      "eval_steps_per_second": 27.858,
+        "f1": 0.948932536293766,
+        "number": 5707,
+        "precision": 0.9255372313843079,
+        "recall": 0.9735412651130191
+      },
+      "eval_loss": 3.1915245056152344,
+      "eval_overall_accuracy": 0.8124138451951584,
+      "eval_overall_f1": 0.8287884657492717,
+      "eval_overall_precision": 0.8006443424777897,
+      "eval_overall_recall": 0.8589833112212835,
+      "eval_runtime": 11.2414,
+      "eval_samples_per_second": 756.666,
+      "eval_steps_per_second": 23.662,
       "step": 3000
     },
     {
       "epoch": 0.69,
       "learning_rate": 3.845113178908467e-05,
-      "loss": 3.1407,
+      "loss": 3.2244,
       "step": 3500
     },
     {
       "epoch": 0.79,
       "learning_rate": 3.680129347323962e-05,
-      "loss": 3.073,
+      "loss": 3.1122,
       "step": 4000
     },
     {
       "epoch": 0.79,
       "eval_COMMENT": {
-        "f1": 0.45161290322580644,
-        "number": 34,
-        "precision": 0.5,
-        "recall": 0.4117647058823529
+        "f1": 0.6578412740022282,
+        "number": 6922,
+        "precision": 0.6020151133501259,
+        "recall": 0.7250794568043918
       },
       "eval_NAME": {
-        "f1": 0.6545454545454545,
-        "number": 26,
-        "precision": 0.6206896551724138,
-        "recall": 0.6923076923076923
+        "f1": 0.8043938887030222,
+        "number": 8833,
+        "precision": 0.7925502692011867,
+        "recall": 0.8165968527114231
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9807353044950956,
+        "number": 7092,
+        "precision": 0.9816358242689646,
+        "recall": 0.9798364354201917
       },
       "eval_RANGE_END": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.6755555555555556,
+        "number": 88,
+        "precision": 0.5547445255474452,
+        "recall": 0.8636363636363636
       },
       "eval_UNIT": {
-        "f1": 0.9500000000000001,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9047619047619048
-      },
-      "eval_loss": 2.9998066425323486,
-      "eval_overall_accuracy": 0.7401960784313726,
-      "eval_overall_f1": 0.736842105263158,
-      "eval_overall_precision": 0.7549019607843137,
-      "eval_overall_recall": 0.719626168224299,
-      "eval_runtime": 8.6908,
-      "eval_samples_per_second": 978.735,
-      "eval_steps_per_second": 30.607,
+        "f1": 0.9479572808950669,
+        "number": 5707,
+        "precision": 0.9180758496141849,
+        "recall": 0.9798493078675311
+      },
+      "eval_loss": 3.055974245071411,
+      "eval_overall_accuracy": 0.8183054176029609,
+      "eval_overall_f1": 0.8368498156162123,
+      "eval_overall_precision": 0.8082289803220036,
+      "eval_overall_recall": 0.8675720969206061,
+      "eval_runtime": 13.0303,
+      "eval_samples_per_second": 652.785,
+      "eval_steps_per_second": 20.414,
       "step": 4000
     },
     {
       "epoch": 0.89,
       "learning_rate": 3.515145515739457e-05,
-      "loss": 3.0814,
+      "loss": 3.0919,
       "step": 4500
     },
     {
       "epoch": 0.99,
       "learning_rate": 3.3501616841549535e-05,
-      "loss": 2.954,
+      "loss": 3.074,
       "step": 5000
     },
     {
       "epoch": 0.99,
       "eval_COMMENT": {
-        "f1": 0.6333333333333334,
-        "number": 34,
-        "precision": 0.7307692307692307,
-        "recall": 0.5588235294117647
+        "f1": 0.6620227729403884,
+        "number": 6922,
+        "precision": 0.6171328671328671,
+        "recall": 0.7139555041895406
       },
       "eval_NAME": {
-        "f1": 0.7037037037037038,
-        "number": 26,
-        "precision": 0.6785714285714286,
-        "recall": 0.7307692307692307
+        "f1": 0.8096975138896684,
+        "number": 8833,
+        "precision": 0.8028043623414199,
+        "recall": 0.816710064530737
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9794429242966393,
+        "number": 7092,
+        "precision": 0.9747242005306521,
+        "recall": 0.9842075578116187
       },
       "eval_RANGE_END": {
-        "f1": 0.5,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 0.3333333333333333
+        "f1": 0.7076923076923077,
+        "number": 88,
+        "precision": 0.6448598130841121,
+        "recall": 0.7840909090909091
       },
       "eval_UNIT": {
-        "f1": 0.975609756097561,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9523809523809523
-      },
-      "eval_loss": 2.8982722759246826,
-      "eval_overall_accuracy": 0.7941176470588235,
-      "eval_overall_f1": 0.7999999999999999,
-      "eval_overall_precision": 0.8367346938775511,
-      "eval_overall_recall": 0.7663551401869159,
-      "eval_runtime": 10.4306,
-      "eval_samples_per_second": 815.482,
-      "eval_steps_per_second": 25.502,
+        "f1": 0.9461453651089815,
+        "number": 5707,
+        "precision": 0.9168310322156475,
+        "recall": 0.9773961801296653
+      },
+      "eval_loss": 2.949470043182373,
+      "eval_overall_accuracy": 0.820892422153823,
+      "eval_overall_f1": 0.840272597816505,
+      "eval_overall_precision": 0.8166809464179793,
+      "eval_overall_recall": 0.8652677885622513,
+      "eval_runtime": 11.1611,
+      "eval_samples_per_second": 762.109,
+      "eval_steps_per_second": 23.833,
       "step": 5000
     },
     {
       "epoch": 1.09,
       "learning_rate": 3.1851778525704485e-05,
-      "loss": 2.9666,
+      "loss": 2.9268,
       "step": 5500
     },
     {
       "epoch": 1.19,
       "learning_rate": 3.0201940209859435e-05,
-      "loss": 2.8463,
+      "loss": 2.936,
       "step": 6000
     },
     {
       "epoch": 1.19,
       "eval_COMMENT": {
-        "f1": 0.5161290322580646,
-        "number": 34,
-        "precision": 0.5714285714285714,
-        "recall": 0.47058823529411764
+        "f1": 0.6634152949942423,
+        "number": 6922,
+        "precision": 0.6245376865195765,
+        "recall": 0.7074544929211211
       },
       "eval_NAME": {
-        "f1": 0.6666666666666666,
-        "number": 26,
-        "precision": 0.6428571428571429,
-        "recall": 0.6923076923076923
+        "f1": 0.8093132590809874,
+        "number": 8833,
+        "precision": 0.8003099402258136,
+        "recall": 0.81852145363976
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9790689534476724,
+        "number": 7092,
+        "precision": 0.9721951897678298,
+        "recall": 0.9860406091370558
       },
       "eval_RANGE_END": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.7348837209302326,
+        "number": 88,
+        "precision": 0.6220472440944882,
+        "recall": 0.8977272727272727
       },
       "eval_UNIT": {
-        "f1": 0.9500000000000001,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9047619047619048
-      },
-      "eval_loss": 2.808798313140869,
-      "eval_overall_accuracy": 0.7696078431372549,
-      "eval_overall_f1": 0.7596153846153846,
-      "eval_overall_precision": 0.7821782178217822,
-      "eval_overall_recall": 0.7383177570093458,
-      "eval_runtime": 8.5489,
-      "eval_samples_per_second": 994.983,
-      "eval_steps_per_second": 31.115,
+        "f1": 0.9469594594594595,
+        "number": 5707,
+        "precision": 0.9140714169248328,
+        "recall": 0.9823024356053969
+      },
+      "eval_loss": 2.889272451400757,
+      "eval_overall_accuracy": 0.8216855184394887,
+      "eval_overall_f1": 0.8412752246905207,
+      "eval_overall_precision": 0.8178910577683989,
+      "eval_overall_recall": 0.8660358913483696,
+      "eval_runtime": 12.795,
+      "eval_samples_per_second": 664.793,
+      "eval_steps_per_second": 20.789,
       "step": 6000
     },
     {
       "epoch": 1.29,
       "learning_rate": 2.855210189401439e-05,
-      "loss": 2.8023,
+      "loss": 2.8799,
       "step": 6500
     },
     {
       "epoch": 1.39,
       "learning_rate": 2.690226357816934e-05,
-      "loss": 2.7953,
+      "loss": 2.7662,
       "step": 7000
     },
     {
       "epoch": 1.39,
       "eval_COMMENT": {
-        "f1": 0.5333333333333333,
-        "number": 34,
-        "precision": 0.6153846153846154,
-        "recall": 0.47058823529411764
+        "f1": 0.6726807593914097,
+        "number": 6922,
+        "precision": 0.6298537569339385,
+        "recall": 0.7217567177116441
       },
       "eval_NAME": {
-        "f1": 0.6792452830188679,
-        "number": 26,
-        "precision": 0.6666666666666666,
-        "recall": 0.6923076923076923
+        "f1": 0.8074248541947062,
+        "number": 8833,
+        "precision": 0.7999777753083676,
+        "recall": 0.8150118872410279
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9814136862855534,
+        "number": 7092,
+        "precision": 0.9800337457817773,
+        "recall": 0.9827975183305132
       },
       "eval_RANGE_END": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.7358490566037735,
+        "number": 88,
+        "precision": 0.6290322580645161,
+        "recall": 0.8863636363636364
       },
       "eval_UNIT": {
-        "f1": 0.9500000000000001,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9047619047619048
-      },
-      "eval_loss": 2.7728922367095947,
-      "eval_overall_accuracy": 0.7647058823529411,
-      "eval_overall_f1": 0.7707317073170732,
-      "eval_overall_precision": 0.8061224489795918,
-      "eval_overall_recall": 0.7383177570093458,
-      "eval_runtime": 10.0867,
-      "eval_samples_per_second": 843.288,
-      "eval_steps_per_second": 26.371,
+        "f1": 0.9479758974794195,
+        "number": 5707,
+        "precision": 0.9191902567478605,
+        "recall": 0.9786227439985982
+      },
+      "eval_loss": 2.862208366394043,
+      "eval_overall_accuracy": 0.8235360764393753,
+      "eval_overall_f1": 0.8432707820327757,
+      "eval_overall_precision": 0.8209503025493503,
+      "eval_overall_recall": 0.8668389078974932,
+      "eval_runtime": 9.9331,
+      "eval_samples_per_second": 856.327,
+      "eval_steps_per_second": 26.779,
       "step": 7000
     },
     {
       "epoch": 1.48,
       "learning_rate": 2.5252425262324292e-05,
-      "loss": 2.7846,
+      "loss": 2.8259,
       "step": 7500
     },
     {
       "epoch": 1.58,
       "learning_rate": 2.3602586946479245e-05,
-      "loss": 2.7561,
+      "loss": 2.7839,
       "step": 8000
     },
     {
       "epoch": 1.58,
       "eval_COMMENT": {
-        "f1": 0.5901639344262295,
-        "number": 34,
-        "precision": 0.6666666666666666,
-        "recall": 0.5294117647058824
+        "f1": 0.6755974419387412,
+        "number": 6922,
+        "precision": 0.6325475860330266,
+        "recall": 0.7249349898873158
       },
       "eval_NAME": {
-        "f1": 0.6545454545454545,
-        "number": 26,
-        "precision": 0.6206896551724138,
-        "recall": 0.6923076923076923
+        "f1": 0.8115015974440895,
+        "number": 8833,
+        "precision": 0.8036190053285968,
+        "recall": 0.8195403600135854
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9806342969407802,
+        "number": 7092,
+        "precision": 0.975977653631285,
+        "recall": 0.9853355893965031
       },
       "eval_RANGE_END": {
-        "f1": 0.8,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 0.6666666666666666
+        "f1": 0.7389162561576356,
+        "number": 88,
+        "precision": 0.6521739130434783,
+        "recall": 0.8522727272727273
       },
       "eval_UNIT": {
-        "f1": 0.975609756097561,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9523809523809523
-      },
-      "eval_loss": 2.715115785598755,
-      "eval_overall_accuracy": 0.7696078431372549,
-      "eval_overall_f1": 0.7788461538461539,
-      "eval_overall_precision": 0.801980198019802,
-      "eval_overall_recall": 0.7570093457943925,
-      "eval_runtime": 10.2877,
-      "eval_samples_per_second": 826.809,
-      "eval_steps_per_second": 25.856,
+        "f1": 0.9483591961332994,
+        "number": 5707,
+        "precision": 0.9188301018731515,
+        "recall": 0.9798493078675311
+      },
+      "eval_loss": 2.780142307281494,
+      "eval_overall_accuracy": 0.8241969900107634,
+      "eval_overall_f1": 0.8452768729641694,
+      "eval_overall_precision": 0.8221239522143753,
+      "eval_overall_recall": 0.8697716639899449,
+      "eval_runtime": 15.1544,
+      "eval_samples_per_second": 561.289,
+      "eval_steps_per_second": 17.553,
       "step": 8000
     },
     {
       "epoch": 1.68,
       "learning_rate": 2.19527486306342e-05,
-      "loss": 2.733,
+      "loss": 2.7185,
       "step": 8500
     },
     {
       "epoch": 1.78,
       "learning_rate": 2.0302910314789152e-05,
-      "loss": 2.789,
+      "loss": 2.7221,
       "step": 9000
     },
     {
       "epoch": 1.78,
       "eval_COMMENT": {
-        "f1": 0.5666666666666668,
-        "number": 34,
-        "precision": 0.6538461538461539,
-        "recall": 0.5
+        "f1": 0.690339005761758,
+        "number": 6922,
+        "precision": 0.6436781609195402,
+        "recall": 0.7442935567754985
       },
       "eval_NAME": {
-        "f1": 0.6071428571428571,
-        "number": 26,
-        "precision": 0.5666666666666667,
-        "recall": 0.6538461538461539
+        "f1": 0.8162713392303792,
+        "number": 8833,
+        "precision": 0.8124719605204127,
+        "recall": 0.8201064191101551
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9827975183305132,
+        "number": 7092,
+        "precision": 0.9827975183305132,
+        "recall": 0.9827975183305132
       },
       "eval_RANGE_END": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.7336683417085428,
+        "number": 88,
+        "precision": 0.6576576576576577,
+        "recall": 0.8295454545454546
       },
       "eval_UNIT": {
-        "f1": 0.9500000000000001,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9047619047619048
-      },
-      "eval_loss": 2.680718183517456,
-      "eval_overall_accuracy": 0.7598039215686274,
-      "eval_overall_f1": 0.7596153846153846,
-      "eval_overall_precision": 0.7821782178217822,
-      "eval_overall_recall": 0.7383177570093458,
-      "eval_runtime": 8.9388,
-      "eval_samples_per_second": 951.583,
-      "eval_steps_per_second": 29.758,
+        "f1": 0.9494640122511486,
+        "number": 5707,
+        "precision": 0.9227716222920457,
+        "recall": 0.9777466269493604
+      },
+      "eval_loss": 2.7520177364349365,
+      "eval_overall_accuracy": 0.8285401363370282,
+      "eval_overall_f1": 0.850812759300823,
+      "eval_overall_precision": 0.8292674842558834,
+      "eval_overall_recall": 0.8735074366315202,
+      "eval_runtime": 11.6589,
+      "eval_samples_per_second": 729.574,
+      "eval_steps_per_second": 22.815,
       "step": 9000
     },
     {
       "epoch": 1.88,
       "learning_rate": 1.8653071998944105e-05,
-      "loss": 2.741,
+      "loss": 2.8004,
       "step": 9500
     },
     {
       "epoch": 1.98,
       "learning_rate": 1.700323368309906e-05,
-      "loss": 2.6709,
+      "loss": 2.7156,
       "step": 10000
     },
     {
       "epoch": 1.98,
       "eval_COMMENT": {
-        "f1": 0.5333333333333333,
-        "number": 34,
-        "precision": 0.6153846153846154,
-        "recall": 0.47058823529411764
+        "f1": 0.6864177489177489,
+        "number": 6922,
+        "precision": 0.6453828542355635,
+        "recall": 0.7330251372435712
       },
       "eval_NAME": {
-        "f1": 0.6428571428571429,
-        "number": 26,
-        "precision": 0.6,
-        "recall": 0.6923076923076923
+        "f1": 0.8142527960433878,
+        "number": 8833,
+        "precision": 0.8084821428571428,
+        "recall": 0.8201064191101551
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9826589595375722,
+        "number": 7092,
+        "precision": 0.9825204398082887,
+        "recall": 0.9827975183305132
       },
       "eval_RANGE_END": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.7219512195121951,
+        "number": 88,
+        "precision": 0.6324786324786325,
+        "recall": 0.8409090909090909
       },
       "eval_UNIT": {
-        "f1": 0.9500000000000001,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9047619047619048
-      },
-      "eval_loss": 2.645069122314453,
-      "eval_overall_accuracy": 0.7696078431372549,
-      "eval_overall_f1": 0.7596153846153846,
-      "eval_overall_precision": 0.7821782178217822,
-      "eval_overall_recall": 0.7383177570093458,
-      "eval_runtime": 9.7262,
-      "eval_samples_per_second": 874.545,
-      "eval_steps_per_second": 27.349,
+        "f1": 0.9496769806188372,
+        "number": 5707,
+        "precision": 0.9222387320455672,
+        "recall": 0.9787979674084458
+      },
+      "eval_loss": 2.7235608100891113,
+      "eval_overall_accuracy": 0.8264629793983799,
+      "eval_overall_f1": 0.8495539058775454,
+      "eval_overall_precision": 0.8291126620139582,
+      "eval_overall_recall": 0.8710285594581384,
+      "eval_runtime": 10.9292,
+      "eval_samples_per_second": 778.285,
+      "eval_steps_per_second": 24.339,
       "step": 10000
     },
     {
       "epoch": 2.08,
       "learning_rate": 1.535339536725401e-05,
-      "loss": 2.657,
+      "loss": 2.6908,
       "step": 10500
     },
     {
       "epoch": 2.18,
       "learning_rate": 1.3703557051408963e-05,
-      "loss": 2.5318,
+      "loss": 2.6804,
       "step": 11000
     },
     {
       "epoch": 2.18,
       "eval_COMMENT": {
-        "f1": 0.49180327868852464,
-        "number": 34,
-        "precision": 0.5555555555555556,
-        "recall": 0.4411764705882353
+        "f1": 0.6895951823352291,
+        "number": 6922,
+        "precision": 0.6422784494578088,
+        "recall": 0.7444380236925744
       },
       "eval_NAME": {
-        "f1": 0.6785714285714285,
-        "number": 26,
-        "precision": 0.6333333333333333,
-        "recall": 0.7307692307692307
+        "f1": 0.8174098858460327,
+        "number": 8833,
+        "precision": 0.8120670391061453,
+        "recall": 0.8228235027736895
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9829240756421113,
+        "number": 7092,
+        "precision": 0.9837570621468926,
+        "recall": 0.9820924985899605
       },
       "eval_RANGE_END": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.7281553398058253,
+        "number": 88,
+        "precision": 0.635593220338983,
+        "recall": 0.8522727272727273
       },
       "eval_UNIT": {
-        "f1": 0.975609756097561,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9523809523809523
-      },
-      "eval_loss": 2.623105764389038,
-      "eval_overall_accuracy": 0.7745098039215687,
-      "eval_overall_f1": 0.7619047619047619,
-      "eval_overall_precision": 0.7766990291262136,
-      "eval_overall_recall": 0.7476635514018691,
-      "eval_runtime": 10.355,
-      "eval_samples_per_second": 821.438,
-      "eval_steps_per_second": 25.688,
+        "f1": 0.9503232391970058,
+        "number": 5707,
+        "precision": 0.9234584228798148,
+        "recall": 0.9787979674084458
+      },
+      "eval_loss": 2.692896604537964,
+      "eval_overall_accuracy": 0.8278792227656401,
+      "eval_overall_f1": 0.8510414189120316,
+      "eval_overall_precision": 0.8288219722038385,
+      "eval_overall_recall": 0.8744850219956707,
+      "eval_runtime": 9.5517,
+      "eval_samples_per_second": 890.518,
+      "eval_steps_per_second": 27.848,
       "step": 11000
     },
     {
       "epoch": 2.28,
       "learning_rate": 1.2053718735563915e-05,
-      "loss": 2.5907,
+      "loss": 2.5859,
       "step": 11500
     },
     {
       "epoch": 2.38,
       "learning_rate": 1.0403880419718868e-05,
-      "loss": 2.6576,
+      "loss": 2.6121,
       "step": 12000
     },
     {
       "epoch": 2.38,
       "eval_COMMENT": {
-        "f1": 0.5999999999999999,
-        "number": 34,
-        "precision": 0.6923076923076923,
-        "recall": 0.5294117647058824
+        "f1": 0.691554715452643,
+        "number": 6922,
+        "precision": 0.6490939044481054,
+        "recall": 0.7399595492632187
       },
       "eval_NAME": {
-        "f1": 0.7169811320754716,
-        "number": 26,
-        "precision": 0.7037037037037037,
-        "recall": 0.7307692307692307
+        "f1": 0.8156028368794326,
+        "number": 8833,
+        "precision": 0.811037725288257,
+        "recall": 0.820219630929469
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9826362038664324,
+        "number": 7092,
+        "precision": 0.9798121407542408,
+        "recall": 0.9854765933446137
       },
       "eval_RANGE_END": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.7499999999999999,
+        "number": 88,
+        "precision": 0.6328125,
+        "recall": 0.9204545454545454
       },
       "eval_UNIT": {
-        "f1": 0.975609756097561,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9523809523809523
-      },
-      "eval_loss": 2.602302074432373,
-      "eval_overall_accuracy": 0.7843137254901961,
-      "eval_overall_f1": 0.8058252427184466,
-      "eval_overall_precision": 0.8383838383838383,
-      "eval_overall_recall": 0.7757009345794392,
-      "eval_runtime": 9.1898,
-      "eval_samples_per_second": 925.595,
-      "eval_steps_per_second": 28.945,
+        "f1": 0.9497709146444936,
+        "number": 5707,
+        "precision": 0.9207106431978944,
+        "recall": 0.9807254249167688
+      },
+      "eval_loss": 2.669058322906494,
+      "eval_overall_accuracy": 0.8276526238268784,
+      "eval_overall_f1": 0.8514097200965888,
+      "eval_overall_precision": 0.8299297175440923,
+      "eval_overall_recall": 0.8740311430766008,
+      "eval_runtime": 10.4859,
+      "eval_samples_per_second": 811.181,
+      "eval_steps_per_second": 25.367,
       "step": 12000
     },
     {
       "epoch": 2.47,
       "learning_rate": 8.75404210387382e-06,
-      "loss": 2.6148,
+      "loss": 2.6392,
       "step": 12500
     },
     {
       "epoch": 2.57,
       "learning_rate": 7.104203788028774e-06,
-      "loss": 2.5836,
+      "loss": 2.553,
       "step": 13000
     },
     {
       "epoch": 2.57,
       "eval_COMMENT": {
-        "f1": 0.576271186440678,
-        "number": 34,
-        "precision": 0.68,
-        "recall": 0.5
+        "f1": 0.6915875260995488,
+        "number": 6922,
+        "precision": 0.6478233438485804,
+        "recall": 0.7416931522681306
       },
       "eval_NAME": {
-        "f1": 0.6909090909090909,
-        "number": 26,
-        "precision": 0.6551724137931034,
-        "recall": 0.7307692307692307
+        "f1": 0.8171171171171171,
+        "number": 8833,
+        "precision": 0.8128150554497592,
+        "recall": 0.8214649609419223
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9832299887260428,
+        "number": 7092,
+        "precision": 0.9826760563380281,
+        "recall": 0.983784545967287
       },
       "eval_RANGE_END": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.742857142857143,
+        "number": 88,
+        "precision": 0.639344262295082,
+        "recall": 0.8863636363636364
       },
       "eval_UNIT": {
-        "f1": 0.975609756097561,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9523809523809523
-      },
-      "eval_loss": 2.574141025543213,
-      "eval_overall_accuracy": 0.7892156862745098,
-      "eval_overall_f1": 0.7922705314009661,
-      "eval_overall_precision": 0.82,
-      "eval_overall_recall": 0.7663551401869159,
-      "eval_runtime": 9.1902,
-      "eval_samples_per_second": 925.552,
-      "eval_steps_per_second": 28.944,
+        "f1": 0.9500635862653668,
+        "number": 5707,
+        "precision": 0.920335085413929,
+        "recall": 0.9817767653758542
+      },
+      "eval_loss": 2.6652376651763916,
+      "eval_overall_accuracy": 0.8286912022962026,
+      "eval_overall_f1": 0.8519148357254608,
+      "eval_overall_precision": 0.8304489092235263,
+      "eval_overall_recall": 0.874519935758676,
+      "eval_runtime": 11.1046,
+      "eval_samples_per_second": 765.987,
+      "eval_steps_per_second": 23.954,
       "step": 13000
     },
     {
       "epoch": 2.67,
       "learning_rate": 5.4543654721837265e-06,
-      "loss": 2.5413,
+      "loss": 2.5782,
       "step": 13500
     },
     {
       "epoch": 2.77,
       "learning_rate": 3.804527156338679e-06,
-      "loss": 2.6004,
+      "loss": 2.5781,
       "step": 14000
     },
     {
       "epoch": 2.77,
       "eval_COMMENT": {
-        "f1": 0.5666666666666668,
-        "number": 34,
-        "precision": 0.6538461538461539,
-        "recall": 0.5
+        "f1": 0.6930304873926858,
+        "number": 6922,
+        "precision": 0.6512514292974209,
+        "recall": 0.7405374169315226
       },
       "eval_NAME": {
-        "f1": 0.6428571428571429,
-        "number": 26,
-        "precision": 0.6,
-        "recall": 0.6923076923076923
+        "f1": 0.8163609767075503,
+        "number": 8833,
+        "precision": 0.8114304887596465,
+        "recall": 0.8213517491226084
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9833638798815735,
+        "number": 7092,
+        "precision": 0.9832252607837609,
+        "recall": 0.983502538071066
       },
       "eval_RANGE_END": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.7450980392156864,
+        "number": 88,
+        "precision": 0.6551724137931034,
+        "recall": 0.8636363636363636
       },
       "eval_UNIT": {
-        "f1": 0.9500000000000001,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9047619047619048
-      },
-      "eval_loss": 2.5654454231262207,
-      "eval_overall_accuracy": 0.7745098039215687,
-      "eval_overall_f1": 0.7692307692307693,
-      "eval_overall_precision": 0.7920792079207921,
-      "eval_overall_recall": 0.7476635514018691,
-      "eval_runtime": 10.3314,
-      "eval_samples_per_second": 823.318,
-      "eval_steps_per_second": 25.747,
+        "f1": 0.9503184713375797,
+        "number": 5707,
+        "precision": 0.9220500988793672,
+        "recall": 0.9803749780970737
+      },
+      "eval_loss": 2.6431026458740234,
+      "eval_overall_accuracy": 0.8285967860717186,
+      "eval_overall_f1": 0.8522440918068515,
+      "eval_overall_precision": 0.8317381189764042,
+      "eval_overall_recall": 0.8737867467355631,
+      "eval_runtime": 10.2487,
+      "eval_samples_per_second": 829.962,
+      "eval_steps_per_second": 25.955,
       "step": 14000
     },
     {
       "epoch": 2.87,
       "learning_rate": 2.154688840493632e-06,
-      "loss": 2.5827,
+      "loss": 2.5714,
       "step": 14500
     },
     {
       "epoch": 2.97,
       "learning_rate": 5.048505246485845e-07,
-      "loss": 2.5449,
+      "loss": 2.5928,
       "step": 15000
     },
     {
       "epoch": 2.97,
       "eval_COMMENT": {
-        "f1": 0.6101694915254237,
-        "number": 34,
-        "precision": 0.72,
-        "recall": 0.5294117647058824
+        "f1": 0.6959701997968167,
+        "number": 6922,
+        "precision": 0.6551064643631264,
+        "recall": 0.7422710199364345
       },
       "eval_NAME": {
-        "f1": 0.6545454545454545,
-        "number": 26,
-        "precision": 0.6206896551724138,
-        "recall": 0.6923076923076923
+        "f1": 0.8171788810086682,
+        "number": 8833,
+        "precision": 0.8126049479458188,
+        "recall": 0.8218045963998641
       },
       "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9831607130275488,
+        "number": 7092,
+        "precision": 0.9825376707505985,
+        "recall": 0.983784545967287
       },
       "eval_RANGE_END": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.7439613526570048,
+        "number": 88,
+        "precision": 0.6470588235294118,
+        "recall": 0.875
       },
       "eval_UNIT": {
-        "f1": 0.9500000000000001,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9047619047619048
-      },
-      "eval_loss": 2.5585994720458984,
-      "eval_overall_accuracy": 0.7892156862745098,
-      "eval_overall_f1": 0.7864077669902914,
-      "eval_overall_precision": 0.8181818181818182,
-      "eval_overall_recall": 0.7570093457943925,
-      "eval_runtime": 9.9625,
-      "eval_samples_per_second": 853.8,
-      "eval_steps_per_second": 26.7,
+        "f1": 0.9497833290848839,
+        "number": 5707,
+        "precision": 0.9219729462223688,
+        "recall": 0.9793236376379885
+      },
+      "eval_loss": 2.639442205429077,
+      "eval_overall_accuracy": 0.8289178012349642,
+      "eval_overall_f1": 0.8531516183986372,
+      "eval_overall_precision": 0.8330560915563244,
+      "eval_overall_recall": 0.874240625654633,
+      "eval_runtime": 9.6239,
+      "eval_samples_per_second": 883.843,
+      "eval_steps_per_second": 27.64,
       "step": 15000
     },
     {
       "epoch": 3.0,
       "step": 15153,
-      "total_flos": 467272509509796.0,
-      "train_loss": 3.027484999280129,
-      "train_runtime": 1025.3838,
-      "train_samples_per_second": 472.807,
-      "train_steps_per_second": 14.778
+      "total_flos": 466306211114724.0,
+      "train_loss": 3.047607653480721,
+      "train_runtime": 1056.0535,
+      "train_samples_per_second": 459.076,
+      "train_steps_per_second": 14.349
     }
   ],
   "logging_steps": 500,
   "max_steps": 15153,
   "num_train_epochs": 3,
   "save_steps": 500,
-  "total_flos": 467272509509796.0,
+  "total_flos": 466306211114724.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:823747ab7556f15a23f379495025c4b71556b054ec3c8b4024c477c4081cb53c
+oid sha256:445f885f648d22fc4ab9f3caa1ac96f712e52c6cede27f3855cfdc76d2db8dfa
 size 4155

validation_results.json

@@ -1,41 +1,41 @@
 {
     "epoch": 3.0,
     "eval_COMMENT": {
-        "f1": 0.6101694915254237,
-        "number": 34,
-        "precision": 0.72,
-        "recall": 0.5294117647058824
+        "f1": 0.6959761549925485,
+        "number": 6922,
+        "precision": 0.6552295918367347,
+        "recall": 0.7421265530193586
     },
     "eval_NAME": {
-        "f1": 0.6545454545454545,
-        "number": 26,
-        "precision": 0.6206896551724138,
-        "recall": 0.6923076923076923
+        "f1": 0.8171377097173741,
+        "number": 8833,
+        "precision": 0.8127449882405644,
+        "recall": 0.8215781727612362
     },
     "eval_QTY": {
-        "f1": 1.0,
-        "number": 23,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.9830914470903199,
+        "number": 7092,
+        "precision": 0.9823993241340467,
+        "recall": 0.983784545967287
     },
     "eval_RANGE_END": {
-        "f1": 1.0,
-        "number": 3,
-        "precision": 1.0,
-        "recall": 1.0
+        "f1": 0.75,
+        "number": 88,
+        "precision": 0.65,
+        "recall": 0.8863636363636364
     },
     "eval_UNIT": {
-        "f1": 0.9500000000000001,
-        "number": 21,
-        "precision": 1.0,
-        "recall": 0.9047619047619048
+        "f1": 0.9497026338147834,
+        "number": 5707,
+        "precision": 0.9218208807521029,
+        "recall": 0.9793236376379885
     },
-    "eval_loss": 2.5587010383605957,
-    "eval_overall_accuracy": 0.7892156862745098,
-    "eval_overall_f1": 0.7864077669902914,
-    "eval_overall_precision": 0.8181818181818182,
-    "eval_overall_recall": 0.7570093457943925,
-    "eval_runtime": 10.3156,
-    "eval_samples_per_second": 824.575,
-    "eval_steps_per_second": 25.786
+    "eval_loss": 2.639784336090088,
+    "eval_overall_accuracy": 0.8287289687859962,
+    "eval_overall_f1": 0.8531416110126756,
+    "eval_overall_precision": 0.8331004192453584,
+    "eval_overall_recall": 0.8741707981286223,
+    "eval_runtime": 13.2958,
+    "eval_samples_per_second": 639.753,
+    "eval_steps_per_second": 20.006
 }