Training update: 6,893/238,453 rows (2.89%) | +501 new @ 2025-10-23 03:09:37

README.md

@@ -24,7 +24,7 @@ pipeline_tag: fill-mask
 - Model type: fine-tuned lightweight BERT variant  
 - Languages: English & Indonesia
 - Finetuned from: `boltuix/bert-micro`
-- Status: **Early version** — trained on **2.85%** of planned data.
+- Status: **Early version** — trained on **2.89%** of planned data.
 **Model sources**  
 - Base model: [boltuix/bert-micro](https://huggingface.co/boltuix/bert-micro)
 - Data: Cybersecurity Data
@@ -40,7 +40,7 @@ You can use this model to classify cybersecurity-related text — for example, w
 - Not optimized for languages other than English and Indonesian.  
 - 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 (2.85%) 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 (2.89%) 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.
 ## 4. How to Get Started with the Model  
@@ -75,11 +75,11 @@ Since cybersecurity data often contains lengthy alert descriptions and execution
 
 ### Training Data
 - **Total database rows**: 238,453
-- **Rows processed (cumulative)**: 6,793 (2.85%)
-- **Rows in this session**: 100
-- **Training samples (after chunking)**: 100
-- **Training date**: 2025-10-23 03:04:55
+- **Rows processed (cumulative)**: 6,893 (2.89%)
+- **Rows in this session**: 501
+- **Training samples (after chunking)**: 501
+- **Training date**: 2025-10-23 03:09:37
 
 ### Post-Training Metrics
-- **Final training loss**: 0.0000
+- **Final training loss**: 
 - **Rows→Samples ratio**: 1.00x (average chunks per row)

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c79ea1001e9ec2feb829349367b545ab600450341050ac6fffa7dcbdf63cc316
+oid sha256:e2e7529b5d461e6e4fab43e90df77f11c204ca761044372d623b3b0399d2206e
 size 17671560

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a0b37e1532e141eddb8c79aca9380dd95f3736c148ba09cab85888ff07a0a34
+oid sha256:c4c01b98595a6b75c1defe26eb86f4938dde769ad3f6e9d3e9be5711880ffb56
 size 5905

training_metadata.json

@@ -1,11 +1,11 @@
 {
-  "trained_at": 1761188695.8412066,
-  "trained_at_readable": "2025-10-23 03:04:55",
-  "samples_this_session": 100,
-  "new_rows_this_session": 100,
-  "trained_rows_total": 6793,
+  "trained_at": 1761188977.395925,
+  "trained_at_readable": "2025-10-23 03:09:37",
+  "samples_this_session": 501,
+  "new_rows_this_session": 501,
+  "trained_rows_total": 6893,
   "total_db_rows": 238453,
-  "percentage": 2.8487794240374416,
+  "percentage": 2.890716409523051,
   "final_loss": 0,
   "epochs": 3,
   "learning_rate": 5e-05,