Hugging Face's logo

ctrltokyo
/
ColBERT-Zero-6L-CodeSearch

Sentence Similarity
Safetensors
sentence-transformers
English
PyLate
modernbert
ColBERT
feature-extraction
code-search
knowledge-distillation
apple-silicon
mps
text-embeddings-inference
Model card Files
xet
 Community
ctrltokyo commited on
Commit
a817643
·
verified ·
1 Parent(s): beaa85c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -39,7 +39,7 @@ A **6-layer ColBERT model** distilled from [ColBERT-Zero](https://huggingface.co
39
 
40
  ## Benchmark Results
41
 
42
- Evaluated on 3 code search corpora (150 questions total) via [litembeddings](https://github.com/nicholasgasior/litembeddings):
43
 
44
  | Corpus | Teacher MRR | Student MRR | % of Teacher | Student Query Speed |
45
  |--------|------------|-------------|--------------|---------------------|
@@ -145,7 +145,7 @@ reranked = rank.rerank(
145
 
146
  ## GGUF / litembeddings
147
 
148
- This model can be converted to GGUF format for use with [litembeddings](https://github.com/nicholasgasior/litembeddings) (SQLite-based embedding engine with SIMD-accelerated MaxSim):
149
 
150
  ```bash
151
  # Convert to GGUF
@@ -191,5 +191,5 @@ SELECT lembed_maxsim(
191
 
192
  - [ColBERT-Zero](https://huggingface.co/lightonai/ColBERT-Zero) by LightOn AI — the teacher model
193
  - [PyLate](https://github.com/lightonai/pylate) — ColBERT training framework
194
- - [litembeddings](https://github.com/nicholasgasior/litembeddings) — SQLite embedding engine used for benchmarking
195
  - Training and experimentation performed entirely on Apple Silicon (M4 Max) using PyTorch MPS backend
 
39
 
40
  ## Benchmark Results
41
 
42
+ Evaluated on 3 code search corpora (150 questions total) via [litembeddings](https://github.com/alexandernicholson/litembeddings):
43
 
44
  | Corpus | Teacher MRR | Student MRR | % of Teacher | Student Query Speed |
45
  |--------|------------|-------------|--------------|---------------------|
 
145
 
146
  ## GGUF / litembeddings
147
 
148
+ This model can be converted to GGUF format for use with [litembeddings](https://github.com/alexandernicholson/litembeddings) (SQLite-based embedding engine with SIMD-accelerated MaxSim):
149
 
150
  ```bash
151
  # Convert to GGUF
 
191
 
192
  - [ColBERT-Zero](https://huggingface.co/lightonai/ColBERT-Zero) by LightOn AI — the teacher model
193
  - [PyLate](https://github.com/lightonai/pylate) — ColBERT training framework
194
+ - [litembeddings](https://github.com/alexandernicholson/litembeddings) — SQLite embedding engine used for benchmarking
195
  - Training and experimentation performed entirely on Apple Silicon (M4 Max) using PyTorch MPS backend