Sentence Similarity
sentence-transformers
Joblib
Safetensors
modernbert
security
intrusion-detection
behavior-analytics
intent-recognition
linux
kubernetes
audit-log
text-embeddings-inference
Instructions to use willchen0011/SecEBL with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use willchen0011/SecEBL with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("willchen0011/SecEBL") sentences = [ "That is a happy person", "That is a happy dog", "That is a very happy person", "Today is a sunny day" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4] - Notebooks
- Google Colab
- Kaggle
Commit ·
b883fb0
1
Parent(s): 697129c
Add public K8s benchmark examples
Browse files- README.md +6 -2
- examples/README.md +7 -2
- examples/k8s/example_gold.rev20.jsonl +3 -0
- examples/k8s/example_sessions.jsonl +3 -0
- examples/manifest.json +14 -0
README.md
CHANGED
|
@@ -84,6 +84,7 @@ This Hugging Face repository is the model artifact bundle.
|
|
| 84 |
| `semantic_texts.jsonl` | Rev20 semantic label texts used by the L1 retrieval path. |
|
| 85 |
| `schema/tags_schema_rev20.json` | Canonical Rev20 behavior vocabulary, 361 tags across 12 groups. |
|
| 86 |
| `examples/linux/` | Public subset of the internal Linux final benchmark and matching Rev20 labels. |
|
|
|
|
| 87 |
| `examples/manifest.json` | Public example subset counts and distribution. |
|
| 88 |
| `rev20_tag_rfc.md` | Rev20 behavior-tag labeling RFC and boundary examples. |
|
| 89 |
| `l2_artifacts/logreg.joblib` | Experimental L2 logistic-regression session scorer. |
|
|
@@ -252,13 +253,16 @@ Hard negatives were designed in two layers:
|
|
| 252 |
|
| 253 |
## Public Benchmark Subset
|
| 254 |
|
| 255 |
-
This Hugging Face repository includes the same public
|
| 256 |
-
|
|
|
|
| 257 |
|
| 258 |
| Public artifact | Rows | Sessions | Notes |
|
| 259 |
| --- | ---: | ---: | --- |
|
| 260 |
| `examples/linux/example_sessions.jsonl` | 10,520 | 531 | Publicly releasable subset of the internal Linux final benchmark; 2,934 normal-operation rows and 7,586 intrusion rows. |
|
| 261 |
| `examples/linux/example_gold.rev20.jsonl` | 10,520 | 531 | Matching Rev20 behavior labels; 10,019 labeled rows, 14,807 behavior-label instances, and 349 unique behavior tags. |
|
|
|
|
|
|
|
| 262 |
|
| 263 |
Session-level labels use English enums: `normal_operation` and `intrusion`.
|
| 264 |
The full internal Linux benchmark remains larger: 12,594 rows, 663 sessions,
|
|
|
|
| 84 |
| `semantic_texts.jsonl` | Rev20 semantic label texts used by the L1 retrieval path. |
|
| 85 |
| `schema/tags_schema_rev20.json` | Canonical Rev20 behavior vocabulary, 361 tags across 12 groups. |
|
| 86 |
| `examples/linux/` | Public subset of the internal Linux final benchmark and matching Rev20 labels. |
|
| 87 |
+
| `examples/k8s/` | Public normalized Kubernetes AuditLog examples and matching Rev20 labels. |
|
| 88 |
| `examples/manifest.json` | Public example subset counts and distribution. |
|
| 89 |
| `rev20_tag_rfc.md` | Rev20 behavior-tag labeling RFC and boundary examples. |
|
| 90 |
| `l2_artifacts/logreg.joblib` | Experimental L2 logistic-regression session scorer. |
|
|
|
|
| 253 |
|
| 254 |
## Public Benchmark Subset
|
| 255 |
|
| 256 |
+
This Hugging Face repository includes the same public benchmark examples as the
|
| 257 |
+
companion GitHub repository: the Linux benchmark subset under `examples/linux/`
|
| 258 |
+
and normalized Kubernetes AuditLog examples under `examples/k8s/`.
|
| 259 |
|
| 260 |
| Public artifact | Rows | Sessions | Notes |
|
| 261 |
| --- | ---: | ---: | --- |
|
| 262 |
| `examples/linux/example_sessions.jsonl` | 10,520 | 531 | Publicly releasable subset of the internal Linux final benchmark; 2,934 normal-operation rows and 7,586 intrusion rows. |
|
| 263 |
| `examples/linux/example_gold.rev20.jsonl` | 10,520 | 531 | Matching Rev20 behavior labels; 10,019 labeled rows, 14,807 behavior-label instances, and 349 unique behavior tags. |
|
| 264 |
+
| `examples/k8s/example_sessions.jsonl` | 144 | 46 | Public normalized Kubernetes AuditLog examples; 72 normal-operation rows and 72 intrusion rows. |
|
| 265 |
+
| `examples/k8s/example_gold.rev20.jsonl` | 144 | 46 | Matching Rev20 behavior labels; 144 labeled rows, 163 behavior-label instances, and 27 unique behavior tags. |
|
| 266 |
|
| 267 |
Session-level labels use English enums: `normal_operation` and `intrusion`.
|
| 268 |
The full internal Linux benchmark remains larger: 12,594 rows, 663 sessions,
|
examples/README.md
CHANGED
|
@@ -1,7 +1,8 @@
|
|
| 1 |
# SecEBL Public Final Benchmark Examples
|
| 2 |
|
| 3 |
-
This directory contains
|
| 4 |
-
|
|
|
|
| 5 |
|
| 6 |
**The full internal benchmark, training corpora, private pressure-stream rows,
|
| 7 |
and removed sensitive benchmark sessions are not distributed in this GitHub
|
|
@@ -15,5 +16,9 @@ subset.
|
|
| 15 |
public subset sessions.
|
| 16 |
- `linux/example_gold.rev20.jsonl`: matching Rev20 behavior tags for the same
|
| 17 |
10,520 rows.
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
|
| 19 |
Session-level labels use English enums: `normal_operation` and `intrusion`.
|
|
|
|
| 1 |
# SecEBL Public Final Benchmark Examples
|
| 2 |
|
| 3 |
+
This directory contains publicly releasable example data for verifying the
|
| 4 |
+
release code path: a subset of the internal Linux final benchmark plus
|
| 5 |
+
normalized Kubernetes AuditLog examples.
|
| 6 |
|
| 7 |
**The full internal benchmark, training corpora, private pressure-stream rows,
|
| 8 |
and removed sensitive benchmark sessions are not distributed in this GitHub
|
|
|
|
| 16 |
public subset sessions.
|
| 17 |
- `linux/example_gold.rev20.jsonl`: matching Rev20 behavior tags for the same
|
| 18 |
10,520 rows.
|
| 19 |
+
- `k8s/example_sessions.jsonl`: 144 normalized Kubernetes AuditLog rows across
|
| 20 |
+
46 sessions.
|
| 21 |
+
- `k8s/example_gold.rev20.jsonl`: matching Rev20 behavior tags for the same 144
|
| 22 |
+
K8s rows, with 163 behavior-label instances and 27 unique behavior tags.
|
| 23 |
|
| 24 |
Session-level labels use English enums: `normal_operation` and `intrusion`.
|
examples/k8s/example_gold.rev20.jsonl
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fdf777b5fd1e7e6d4bd85276ce0125ec4f7462d849ac78deea75c73000b16cd8
|
| 3 |
+
size 99593
|
examples/k8s/example_sessions.jsonl
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8913939e98a9d7f09a6eab49a32dd914f5df6e399b3c072e917b489db89a65c2
|
| 3 |
+
size 60154
|
examples/manifest.json
CHANGED
|
@@ -1,4 +1,18 @@
|
|
| 1 |
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
"linux": {
|
| 3 |
"behavior_label_instances": 14807,
|
| 4 |
"files": [
|
|
|
|
| 1 |
{
|
| 2 |
+
"k8s": {
|
| 3 |
+
"behavior_label_instances": 163,
|
| 4 |
+
"files": [
|
| 5 |
+
"examples/k8s/example_sessions.jsonl",
|
| 6 |
+
"examples/k8s/example_gold.rev20.jsonl"
|
| 7 |
+
],
|
| 8 |
+
"intrusion_rows": 72,
|
| 9 |
+
"normal_rows": 72,
|
| 10 |
+
"rows": 144,
|
| 11 |
+
"rows_with_behavior_labels": 144,
|
| 12 |
+
"sessions": 46,
|
| 13 |
+
"source": "public normalized Kubernetes AuditLog examples with public-only metadata",
|
| 14 |
+
"unique_behavior_labels": 27
|
| 15 |
+
},
|
| 16 |
"linux": {
|
| 17 |
"behavior_label_instances": 14807,
|
| 18 |
"files": [
|