Model save

README.md

@@ -21,11 +21,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [ai4bharat/indic-bert](https://huggingface.co/ai4bharat/indic-bert) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1000
-- Precision: 0.8868
-- Recall: 0.8982
-- F1: 0.8925
-- Accuracy: 0.9773
+- Loss: 0.1922
+- Precision: 0.8477
+- Recall: 0.8205
+- F1: 0.8339
+- Accuracy: 0.9481
 
 ## Model description
 
@@ -57,14 +57,14 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.417         | 1.0   | 164  | 0.2686          | 0.7525    | 0.7181 | 0.7349 | 0.9421   |
-| 0.1602        | 2.0   | 328  | 0.1437          | 0.8707    | 0.8667 | 0.8687 | 0.9661   |
-| 0.132         | 3.0   | 492  | 0.1000          | 0.8868    | 0.8982 | 0.8925 | 0.9773   |
+| 0.2768        | 1.0   | 250  | 0.2721          | 0.8527    | 0.7826 | 0.8162 | 0.9334   |
+| 0.1821        | 2.0   | 500  | 0.2106          | 0.8681    | 0.8094 | 0.8378 | 0.9428   |
+| 0.1492        | 3.0   | 750  | 0.1922          | 0.8477    | 0.8205 | 0.8339 | 0.9481   |
 
 
 ### Framework versions
 
 - Transformers 4.57.1
-- Pytorch 2.8.0+cu126
+- Pytorch 2.9.0+cu126
 - Datasets 4.4.1
 - Tokenizers 0.22.1

logs/events.out.tfevents.1763831618.f0da10c9a12b.18129.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5f0c6f1c284b61cc6fa3d4ef40b6b8820d64e99b884c144b05d5138fe683595
+size 10501

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d6333a29dd39833a1d9ff9c0fc0c3a86ef7dfe6606f6634f5fd456607a833137
+oid sha256:1222778c3f0aaf23381a5d8e6ddfca624dd8ef82b43a9c39c4765ea1b1284c9b
 size 131430636