Training update: 1,398/237,619 rows (0.59%) | +100 new @ 2025-10-20 06:10:12

README.md

@@ -18,7 +18,7 @@ base_model: boltuix/bert-micro
 - Model type: fine-tuned lightweight BERT variant  
 - Languages: English & Indonesia
 - Finetuned from: `boltuix/bert-micro`
-- Status: **Early version** — trained on **0.55%** of planned data.
+- Status: **Early version** — trained on **0.59%** of planned data.
 
 **Model sources**  
 - Base model: [boltuix/bert-micro](https://huggingface.co/boltuix/bert-micro)
@@ -41,7 +41,7 @@ You can use this model to classify cybersecurity-related text — for example, w
 
 ## 3. Bias, Risks, and Limitations 
 
-Because the model is based on a small subset (0.55%) 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 (0.59%) 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.
@@ -63,7 +63,7 @@ predicted_class = logits.argmax(dim=-1).item()
 
 ## 5. Training Details
 
-- **Trained records**: 1,298 / 237,619 (0.55%)
+- **Trained records**: 1,398 / 237,619 (0.59%)
 - **Learning rate**: 5e-05
 - **Epochs**: 3
 - **Batch size**: 1

checkpoint-300/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b85317e9a2f99711546b6e16863386c4a09af95d204635546723ec5e2ede845
+oid sha256:c2cdb393c590e97126f470f40c54d65ffefdb8bbe59e24272f3416693d1a4986
 size 17671560

checkpoint-300/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba5c8d507f60f14b58e5d1ddec62c5f8925f5ebc742b3719184ca90f50388344
+oid sha256:6ee6a772c5b151fda8641997e10050e70ce54095047c68b8d047a04669e97a96
 size 35368075

checkpoint-300/trainer_state.json

@@ -11,23 +11,23 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 38.67572784423828,
+      "grad_norm": 39.51091766357422,
       "learning_rate": 3.563829787234043e-05,
-      "loss": 3.1389,
+      "loss": 2.526,
       "step": 100
     },
     {
       "epoch": 2.0,
-      "grad_norm": 57.85369110107422,
+      "grad_norm": 40.35090255737305,
       "learning_rate": 1.7907801418439718e-05,
-      "loss": 2.5113,
+      "loss": 2.1323,
       "step": 200
     },
     {
       "epoch": 3.0,
-      "grad_norm": 76.84485626220703,
+      "grad_norm": 15.34549331665039,
       "learning_rate": 1.773049645390071e-07,
-      "loss": 2.1936,
+      "loss": 1.8826,
       "step": 300
     }
   ],

checkpoint-300/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d463622f4d220e0c62d2a0a1f79afb5f4bee9c1831f3af26ac9165ffc83278bb
+oid sha256:141535d1c3044e612f8853b14d81e1fa4aaa4e2439becfdb8741a6cb1f511a76
 size 5841

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d5c7b02bb13ebf2905410619055cb013385b4011c032639fe17fcccdebaa0782
+oid sha256:c2cdb393c590e97126f470f40c54d65ffefdb8bbe59e24272f3416693d1a4986
 size 17671560

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:923e344d96e07850933e9f1da23fb1dd6dd58910ee71a5248d69e14f54590898
+oid sha256:141535d1c3044e612f8853b14d81e1fa4aaa4e2439becfdb8741a6cb1f511a76
 size 5841

training_metadata.json

@@ -1,11 +1,11 @@
 {
-  "trained_at": 1760940081.235496,
-  "trained_at_readable": "2025-10-20 06:01:21",
-  "samples_this_session": 133,
-  "new_rows_this_session": 50,
-  "trained_rows_total": 1298,
+  "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": 0.5462526144794818,
+  "percentage": 0.5883367912498579,
   "final_loss": 0,
   "epochs": 3,
   "learning_rate": 5e-05