Add model_max_length (32768) to YAML and Model Details section

#6
Files changed (1) hide show
  1. README.md +19 -12
README.md CHANGED
@@ -12,6 +12,7 @@ tags:
12
  - stem
13
  - medical
14
  library_name: sentence-transformers
 
15
  ---
16
 
17
  <img src="https://i.imgur.com/oxvhvQu.png"/>
@@ -30,6 +31,15 @@ This model is released under a non-commercial license. If you'd like a commercia
30
 
31
  For this model's smaller twin, see [zerank-1-small](https://huggingface.co/zeroentropy/zerank-1-small), which we've fully open-sourced under an Apache 2.0 License.
32
 
 
 
 
 
 
 
 
 
 
33
  ## How to Use
34
 
35
  ```python
@@ -43,7 +53,6 @@ query_documents = [
43
  ]
44
 
45
  scores = model.predict(query_documents)
46
-
47
  print(scores)
48
  ```
49
 
@@ -53,15 +62,13 @@ The model can also be inferenced using ZeroEntropy's [/models/rerank](https://do
53
 
54
  NDCG@10 scores between `zerank-1` and competing closed-source proprietary rerankers. Since we are evaluating rerankers, OpenAI's `text-embedding-3-small` is used as an initial retriever for the Top 100 candidate documents.
55
 
56
- | Task | Embedding | cohere-rerank-v3.5 | Salesforce/Llama-rank-v1 | zerank-1-small | **zerank-1** |
57
  |----------------|-----------|--------------------|--------------------------|----------------|--------------|
58
- | Code | 0.678 | 0.724 | 0.694 | 0.730 | **0.754** |
59
- | Conversational | 0.250 | 0.571 | 0.484 | 0.556 | **0.596** |
60
- | Finance | 0.839 | 0.824 | 0.828 | 0.861 | **0.894** |
61
- | Legal | 0.703 | 0.804 | 0.767 | 0.817 | **0.821** |
62
- | Medical | 0.619 | 0.750 | 0.719 | 0.773 | **0.796** |
63
- | STEM | 0.401 | 0.510 | 0.595 | 0.680 | **0.694** |
64
-
65
- Comparing BM25 and Hybrid Search without and with zerank-1:
66
-
67
- <img src="https://cdn-uploads.huggingface.co/production/uploads/67776f9dcd9c9435499eafc8/2GPVHFrI39FspnSNklhsM.png" alt="Description" width="400"/> <img src="https://cdn-uploads.huggingface.co/production/uploads/67776f9dcd9c9435499eafc8/dwYo2D7hoL8QiE8u3yqr9.png" alt="Description" width="400"/>
 
12
  - stem
13
  - medical
14
  library_name: sentence-transformers
15
+ model_max_length: 32768
16
  ---
17
 
18
  <img src="https://i.imgur.com/oxvhvQu.png"/>
 
31
 
32
  For this model's smaller twin, see [zerank-1-small](https://huggingface.co/zeroentropy/zerank-1-small), which we've fully open-sourced under an Apache 2.0 License.
33
 
34
+ ## Model Details
35
+
36
+ | Property | Value |
37
+ |---|---|
38
+ | Parameters | 4B |
39
+ | Context Length | 32,768 tokens (32k) |
40
+ | Base Model | Qwen/Qwen3-4B |
41
+ | License | CC-BY-NC-4.0 |
42
+
43
  ## How to Use
44
 
45
  ```python
 
53
  ]
54
 
55
  scores = model.predict(query_documents)
 
56
  print(scores)
57
  ```
58
 
 
62
 
63
  NDCG@10 scores between `zerank-1` and competing closed-source proprietary rerankers. Since we are evaluating rerankers, OpenAI's `text-embedding-3-small` is used as an initial retriever for the Top 100 candidate documents.
64
 
65
+ | Task | Embedding | cohere-rerank-v3.5 | Salesforce/Llama-rank-v1 | zerank-1-small | **zerank-1** |
66
  |----------------|-----------|--------------------|--------------------------|----------------|--------------|
67
+ | Code | 0.678 | 0.724 | 0.694 | 0.730 | **0.754** |
68
+ | Conversational | 0.250 | 0.571 | 0.484 | 0.556 | **0.596** |
69
+ | Finance | 0.839 | 0.824 | 0.828 | 0.861 | **0.894** |
70
+ | Legal | 0.703 | 0.804 | 0.767 | 0.817 | **0.821** |
71
+ | Medical | 0.619 | 0.750 | 0.719 | 0.773 | **0.796** |
72
+ | STEM | 0.401 | 0.510 | 0.595 | 0.680 | **0.694** |
73
+
74
+ Comparing BM25 and Hybrid Search without and with zerank-1: Description Description