lfoppiano commited on
Commit
d5cc518
2 Parent(s): 4037844 c5c0bc4

Merge branch 'main' into pdf-render

Browse files
Files changed (2) hide show
  1. README.md +1 -1
  2. streamlit_app.py +1 -1
README.md CHANGED
@@ -42,7 +42,7 @@ Allow to change the number of blocks from the original document that are conside
42
  The default size of each block is 250 tokens (which can be changed before uploading the first document).
43
  With default settings, each question uses around 1000 tokens.
44
 
45
- **NOTE**: if the chat answers something like "the information is not provided in the given context", **changing the context size might be a solution**~~~~.
46
 
47
  ### Chunks size
48
  When uploaded, each document is split into blocks of a determined size (250 tokens by default).
 
42
  The default size of each block is 250 tokens (which can be changed before uploading the first document).
43
  With default settings, each question uses around 1000 tokens.
44
 
45
+ **NOTE**: if the chat answers something like "the information is not provided in the given context", **changing the context size will likely help**.
46
 
47
  ### Chunks size
48
  When uploaded, each document is split into blocks of a determined size (250 tokens by default).
streamlit_app.py CHANGED
@@ -163,7 +163,7 @@ with st.sidebar:
163
  st.session_state['model'] = model = st.radio(
164
  "Model",
165
  ("chatgpt-3.5-turbo", "mistral-7b-instruct-v0.1", "zephyr-7b-beta"),
166
- index=1,
167
  captions=[
168
  "ChatGPT 3.5 Turbo + Ada-002-text (embeddings)",
169
  "Mistral-7B-Instruct-V0.1 + Sentence BERT (embeddings) :free:",
 
163
  st.session_state['model'] = model = st.radio(
164
  "Model",
165
  ("chatgpt-3.5-turbo", "mistral-7b-instruct-v0.1", "zephyr-7b-beta"),
166
+ index=2,
167
  captions=[
168
  "ChatGPT 3.5 Turbo + Ada-002-text (embeddings)",
169
  "Mistral-7B-Instruct-V0.1 + Sentence BERT (embeddings) :free:",