Training update: 3,116/237,957 rows (1.31%) | +100 new @ 2025-10-21 11:48:19

README.md

@@ -9,13 +9,8 @@ tags:
 - cybersecurity
 - fill-mask
 - named-entity-recognition
-- transformers
-- tensorflow
-- pytorch
-- masked-language-modeling
 base_model: boltuix/bert-micro
 library_name: transformers
-pipeline_tag: fill-mask
 ---
 # bert-micro-cybersecurity
 
@@ -26,7 +21,7 @@ pipeline_tag: fill-mask
 - Model type: fine-tuned lightweight BERT variant  
 - Languages: English & Indonesia
 - Finetuned from: `boltuix/bert-micro`
-- Status: **Early version** — trained on **1.27%** of planned data.
+- Status: **Early version** — trained on **1.31%** of planned data.
 
 **Model sources**  
 - Base model: [boltuix/bert-micro](https://huggingface.co/boltuix/bert-micro)
@@ -47,7 +42,7 @@ You can use this model to classify cybersecurity-related text — for example, w
 - Not tested for non-cybersecurity domains or out-of-distribution data.
 
 ## 3. Bias, Risks, and Limitations 
-Because the model is based on a small subset (1.27%) of planned data, performance is preliminary and may degrade on unseen or specialized domains (industrial control, IoT logs, foreign language).
+Because the model is based on a small subset (1.31%) of planned data, performance is preliminary and may degrade on unseen or specialized domains (industrial control, IoT logs, foreign language).
 
 - Inherits any biases present in the base model (`boltuix/bert-micro`) and in the fine-tuning data — e.g., over-representation of certain threat types, vendor or tooling-specific vocabulary.
 - Should not be used as sole authority for incident decisions; only as an aid to human analysts.
@@ -67,7 +62,7 @@ predicted_class = logits.argmax(dim=-1).item()
 ```
 
 ## 5. Training Details
-- **Trained records**: 3,016 / 237,957 (1.27%)
+- **Trained records**: 3,116 / 237,957 (1.31%)
 - **Learning rate**: 5e-05
 - **Epochs**: 3
 - **Batch size**: 16

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f1b6a9ba7dbef061ef3a4a4f85937601b1f6f598bd789df8e7804ad3c3425bb8
+oid sha256:583f4fb0896e13bba170d0e0dbf2470093c6b803cbb1aace70868eace83b7cc3
 size 17671560

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b33f116a5a4c8825de97e08426b77f51aece13c2ebe9394937fe22865e4eb7ae
+oid sha256:0698d5f1ca4096f87970681a4fc99b8eb8828f6b682761a992c2b75d3a9ffeb4
 size 5905

training_metadata.json

@@ -1,11 +1,11 @@
 {
-  "trained_at": 1761046953.984294,
-  "trained_at_readable": "2025-10-21 11:42:33",
-  "samples_this_session": 343,
+  "trained_at": 1761047299.4826186,
+  "trained_at_readable": "2025-10-21 11:48:19",
+  "samples_this_session": 400,
   "new_rows_this_session": 100,
-  "trained_rows_total": 3016,
+  "trained_rows_total": 3116,
   "total_db_rows": 237957,
-  "percentage": 1.2674558848867652,
+  "percentage": 1.3094802842530373,
   "final_loss": 0,
   "epochs": 3,
   "learning_rate": 5e-05