Trying to start a new chat in "instruct" mode with Llama 3 gives error and doesn't start the new chat - is there a way to fix?

[TechScorpion]

Today I updated to the latest oobabooga text gen version (despite the issue in my last post about it seeming to contact the internet while processing prompts for some unknown reason - which seems a security risk).

So I tried using another AI model then loaded Llama 3 (Meta-Llama-3-70B-Instruct.IQ1_M.gguf - one of the most quantized versions of it, which was loaded with the llama.cpp model loader, which I assume was the default for it) and tried to start a new conversation using the "new conversation" button and with it in instruct mode.

and it gave this error (in Windows 10 Pro):

IndexError: list index out of range
Traceback (most recent call last):
  File "C:\progs\text-generation-webui\text-generation-webui\installer_files\env\Lib\site-packages\gradio\queueing.py", line 527, in process_events
    response = await route_utils.call_process_api(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\progs\text-generation-webui\text-generation-webui\installer_files\env\Lib\site-packages\gradio\route_utils.py", line 261, in call_process_api
    output = await app.get_blocks().process_api(
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\progs\text-generation-webui\text-generation-webui\installer_files\env\Lib\site-packages\gradio\blocks.py", line 1786, in process_api
    result = await self.call_function(
             ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\progs\text-generation-webui\text-generation-webui\installer_files\env\Lib\site-packages\gradio\blocks.py", line 1338, in call_function
    prediction = await anyio.to_thread.run_sync(
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\progs\text-generation-webui\text-generation-webui\installer_files\env\Lib\site-packages\anyio\to_thread.py", line 33, in run_sync
    return await get_asynclib().run_sync_in_worker_thread(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\progs\text-generation-webui\text-generation-webui\installer_files\env\Lib\site-packages\anyio\_backends\_asyncio.py", line 877, in run_sync_in_worker_thread
    return await future
           ^^^^^^^^^^^^
  File "C:\progs\text-generation-webui\text-generation-webui\installer_files\env\Lib\site-packages\anyio\_backends\_asyncio.py", line 807, in run
    result = context.run(func, *args)
             ^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\progs\text-generation-webui\text-generation-webui\installer_files\env\Lib\site-packages\gradio\utils.py", line 759, in wrapper
    response = f(*args, **kwargs)
               ^^^^^^^^^^^^^^^^^^
  File "C:\progs\text-generation-webui\text-generation-webui\modules\chat.py", line 1052, in handle_start_new_chat_click
    histories = find_all_histories_with_first_prompts(state)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\progs\text-generation-webui\text-generation-webui\modules\chat.py", line 585, in find_all_histories_with_first_prompts
    if data['internal'][0][0] == '<|BEGIN-VISIBLE-CHAT|>':
       ~~~~~~~~~~~~~~~~^^^
**IndexError: list index out of range**

And won't start the new chat. Is the a bug in the latest version of it or is there something I can do to make it work?

[TechScorpion]

I also updated to the new latest version of textgen webui today (25/07/2024) since it says version v1.12 was released today. I assume that's the one currently installed now (though I don't see the version in the UI).

With this latest version it still gives an error of "IndexError: list index out of range" when pressing the button to start a new chat when in Instruct mode when Meta-Llama-3.1-8B-Instruct.Q4_0.gguf is loaded with llama.cpp. Yet it works okay if it's pressed when it's in "chat instruct" mode. But sometimes you want the plainer Instruct mode rather than a character. And Llama.cpp seemed to be the loader that it could load with (I tried a few other ones that gave errors).

Is there fix for pressing "new chat" when in "instruct mode" when using Llama 3 or Llama 3.1 (so it doesn't give that index error when it's pressed)?