Training update: 1,598/237,655 rows (0.67%) | +1 new @ 2025-10-20 09:29:54

README.md

@@ -3,29 +3,31 @@ language:
 - en
 - id
 tags:
+- bert
 - text-classification
+- token-classification
 - cybersecurity
+- fill-mask
+- named-entity-recognition
 base_model: boltuix/bert-micro
+library_name: transformers
 ---
-
 # bert-micro-cybersecurity
 
 ## 1. Model Details
-
 **Model description**  
 "bert-micro-cybersecurity" is a compact transformer model adapted for cybersecurity text classification tasks (e.g., threat detection, incident reports, malicious vs benign content).
 
 - Model type: fine-tuned lightweight BERT variant  
 - Languages: English & Indonesia
 - Finetuned from: `boltuix/bert-micro`
-- Status: **Early version** — trained on **0.59%** of planned data.
+- Status: **Early version** — trained on **0.67%** of planned data.
 
 **Model sources**  
 - Base model: [boltuix/bert-micro](https://huggingface.co/boltuix/bert-micro)
 - Data: Cybersecurity Data
 
 ## 2. Uses
-
 ### Direct use  
 You can use this model to classify cybersecurity-related text — for example, whether a given message, report or log entry indicates malicious intent, abnormal behaviour, or threat presence.
 
@@ -40,14 +42,12 @@ 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 (0.59%) 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 (0.67%) 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  
-
 ```python
 from transformers import AutoTokenizer, AutoModelForSequenceClassification
 
@@ -62,8 +62,7 @@ predicted_class = logits.argmax(dim=-1).item()
 ```
 
 ## 5. Training Details
-
-- **Trained records**: 1,398 / 237,619 (0.59%)
+- **Trained records**: 1,598 / 237,655 (0.67%)
 - **Learning rate**: 5e-05
 - **Epochs**: 3
 - **Batch size**: 1

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c2cdb393c590e97126f470f40c54d65ffefdb8bbe59e24272f3416693d1a4986
+oid sha256:c8256fb20d6b61fe0310f53219033d25608c25418739f63cc54efbe6e6a2f833
 size 17671560

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:141535d1c3044e612f8853b14d81e1fa4aaa4e2439becfdb8741a6cb1f511a76
-size 5841
+oid sha256:944823f3e9b0060cf3527486bbe82b1c34c2c3c5b3e9cefe53fa6360de9c5a72
+size 5905

training_metadata.json

@@ -1,11 +1,11 @@
 {
-  "trained_at": 1760940612.8562534,
-  "trained_at_readable": "2025-10-20 06:10:12",
-  "samples_this_session": 100,
-  "new_rows_this_session": 100,
-  "trained_rows_total": 1398,
-  "total_db_rows": 237619,
-  "percentage": 0.5883367912498579,
+  "trained_at": 1760952594.7217977,
+  "trained_at_readable": "2025-10-20 09:29:54",
+  "samples_this_session": 1,
+  "new_rows_this_session": 1,
+  "trained_rows_total": 1598,
+  "total_db_rows": 237655,
+  "percentage": 0.6724032736529844,
   "final_loss": 0,
   "epochs": 3,
   "learning_rate": 5e-05