Sudehsna commited on
Commit
5ca2779
·
verified ·
1 Parent(s): ac36b8c

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +4 -0
README.md CHANGED
@@ -83,6 +83,7 @@ In total there were three different datasets used for alignment:
83
  **General:**
84
  - Token length limits: Embeddings were truncated at 8192 tokens (OpenAI model constraint)
85
  - Sentence Length Penalties:
 
86
  **Absolute penalty**:
87
  `penalized_sim = cos_sim - α * abs(len_rm - len_de)`
88
 
@@ -182,8 +183,11 @@ parallel_data_unfiltered/
182
  ├── matched_{THRESHOLD}_{ALPHA}_{LENGTH_PENALTY_TYPE}.jsonl
183
 
184
  ├── matched_0.50_ 0.005_relative.jsonl
 
185
  ├── matched_0.50_0.005_absolute.jsonl
 
186
  ├── matched_0.50_none.jsonl
 
187
  ...
188
 
189
  With ALPHA being `0.005` or `None` and LENGTH_PENALTY_TYPE can be `"relative"`, `"absolute"` or `None`.
 
83
  **General:**
84
  - Token length limits: Embeddings were truncated at 8192 tokens (OpenAI model constraint)
85
  - Sentence Length Penalties:
86
+
87
  **Absolute penalty**:
88
  `penalized_sim = cos_sim - α * abs(len_rm - len_de)`
89
 
 
183
  ├── matched_{THRESHOLD}_{ALPHA}_{LENGTH_PENALTY_TYPE}.jsonl
184
 
185
  ├── matched_0.50_ 0.005_relative.jsonl
186
+
187
  ├── matched_0.50_0.005_absolute.jsonl
188
+
189
  ├── matched_0.50_none.jsonl
190
+
191
  ...
192
 
193
  With ALPHA being `0.005` or `None` and LENGTH_PENALTY_TYPE can be `"relative"`, `"absolute"` or `None`.