Skip to content

Commit

Permalink
deploy: 7f78b9a
Browse files Browse the repository at this point in the history
  • Loading branch information
@do-me
do-me committed Mar 4, 2024
1 parent 8d4d773 commit 84fbaf7
Show file tree
Hide file tree
Showing 3 changed files with 111 additions and 7 deletions.
68 changes: 67 additions & 1 deletion bundle.js
View file

Large diffs are not rendered by default.

46 changes: 42 additions & 4 deletions index.html
View file
Original file line number Diff line number Diff line change
Expand Up @@ -706,8 +706,8 @@ <h4>Dimensionality Reduction (New🔥) </h4>
<div class="col-12">
<div>
<hr />
<h4>Chat (Retrieval Augmented Generation, RAG)</h4>
<p>Enter a question to be answered and use the placeholders <b>SEARCH_RESULTS</b> or <b>FULL_TEXT</b> for context.<br> If you encounter errors, the input is probably
<h4>Chat</h4>
<p>Enter a question to be answered and use the placeholders <b>SEARCH_RESULTS</b> or <b>FULL_TEXT</b> for context (Retrieval Augmented Generation, RAG).<br> If you encounter errors, the input is probably
too long (either too many or too
long results or too long prompt). Also, make
sure to check the right prompting style! Xenova/Qwen1.5-1.8B-Chat is by far the best
Expand All @@ -719,7 +719,7 @@ <h4>Chat (Retrieval Augmented Generation, RAG)</h4>
<div class="form-floating input-group mb-2">
<input id="chat_query" class="form-control"
value="Based on the following context, answer the question: What do these paragraphs have in common? Context: SEARCH_RESULTS">
<label for="query-text">Chat Query</label>
<label for="chat_query">Chat Query</label>
</div>

<div class="d-flex flex-row">
Expand Down Expand Up @@ -766,7 +766,7 @@ <h4>Chat (Retrieval Augmented Generation, RAG)</h4>
</div>
<div class="col-md-4">
<div class="form-floating mb-2">
<input type="number" id="chat_max_new_tokens" class="form-control" value="100"
<input type="number" id="chat_max_new_tokens" class="form-control" value="10000"
min="1" step="1" />
<label for="max_new_tokens"># max new tokens</label>
</div>
Expand All @@ -792,7 +792,45 @@ <h4>Chat (Retrieval Augmented Generation, RAG)</h4>

</div>
</div>
</div>
<div class="col-12">
<div>
<hr />
<h4>Ollama Chat Integration</h4>
<p>Enter a question to be answered and use the placeholders <b>SEARCH_RESULTS</b> or <b>FULL_TEXT</b> for context.<br> Install <a href="https://ollama.com/" target="_blank">Ollama</a> locally on macOS, Linux or Windows and connect your server (currently only default http://localhost:11434 supported).</p>
<div class="form-floating input-group mb-2">
<input id="ollama_chat_query" class="form-control"
value="Based on the following context, answer the question: What is this text about? Context: FULL_TEXT">
<label for="ollama_chat_query">Chat Query</label>
</div>

<div class="d-flex flex-row">
<button id="ollama_get_chat" class="btn btn-md btn-primary mb-2 nav-button">🦙 Chat</button>
<div class="row">
<div class="col-md-4" hidden>
<div class="form-floating mb-2">
<input type="text" id="ollama_chat_server" class="form-control" value="http://localhost:11434"/>
<label for="ollama_chat_server">Ollama Server</label>
</div>
</div>
<div class="col-md-8">
<div class="form-floating mb-8">
<input type="text" id="ollama_chat_model" class="form-control" value="llama2"/>
<label for="ollama_chat_model">Model</label>
</div>
</div>
<div class="col-md-4" hidden>
<div class="form-floating mb-2">
<input type="number" id="ollama_chat_max_new_tokens" class="form-control" value="100"
min="1" step="1" />
<label for="ollama_chat_max_new_tokens"># max new tokens</label>
</div>
</div>
</div>

</div>
<div id="ollama_chat_text" class="ml-2"></div>
</div>
</div>
<div class="col-12">
<div>
Expand Down
4 changes: 2 additions & 2 deletions main.css
View file

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

0 comments on commit 84fbaf7

Please sign in to comment.