Here is some info I posted for the 11b version of this model, but it is probably useful for the original 7B version as well.

I think I found the key to avoid repetitions and long rambling answers, which this model has a tendency to do. Hopefully a further fine tune will reduce it. The key is to reduce creativity all the way down, and make the model deterministic. How do you do that?, you may ask. Easy, it is controlled by the following 3 inference parameters: temp, top_p, and top_k

With the following default settings I often get repetitions or additional rambling information:

    "top_k": 40,
    "top_p": 0.95,
    "temp": 0.8,

If I use the following values instead, to make the model deterministic, the problem seems to be gone:

    "top_k": 1,
    "top_p": 0.1,
    "temp": 0.1,

Please note that if you want to use the model for story writing, maybe you get better results by dialing up the creativity.

Here is my complete config file for LM Studio:

{
  "name": "OpenChat",
  "inference_params": {
    "top_k": 1,
    "top_p": 0.1,
    "temp": 0.1,
    "input_prefix": "GPT4 Correct User: ",
    "input_suffix": "<|end_of_turn|>GPT4 Correct Assistant:",
    "antiprompt": [
      "GPT4",
      "<|end_of_turn|>",
      "[End of Turn]",
      "[]"
    ],
    "pre_prompt": "Below is an instruction that describes a task. Write a concise response that appropriately completes the request. Ensure all essential details are provided. Each of your statements must be unique.",
    "pre_prompt_suffix": "<|end_of_turn|>",
    "pre_prompt_prefix": "GPT4 System: "
  }
}

A few words about the above:

• I only include necessary options to avoid overwriting user settings when loading the model or switching prompt format. If you export a config file, please make sure you then edit it manually to clean it up.
• GPT Correct User/Assistant. The Correct keyword is important. It refers to the training data, where the answers were verified as correct. If you do not use it (eg: GPT4 User or Human User), it will still works, but it will give more weight to training data which was unverified.
• GPT4 Sytem or just System are the 2 official recommended ways to prefix system messages. Either work.
• In my system message (pre_promt), I avoid any negative (eg: I do not instruct : Do not repeat yourself"). Remember this is just a language model: if it sees the word “repeat”, it will have a tendency to see this as an instruction to create repetitions! Instead I turned it around into a positive statement based on the word “unique”.

​

As a bonus, here is my config for generating code, which according to my limited testing, this model seems to be surprisingly good at:

{
  "name": "OpenChat Code",
  "inference_params": {
    "top_k": 1,
    "top_p": 0.1,
    "temp": 0.1,
    "input_prefix": "Code User: ",
    "input_suffix": "<|end_of_turn|>Code Assistant:",
    "antiprompt": [
      "GPT4",
      "<|end_of_turn|>",
      "[End of Turn]",
      "[]"
    ],
    "pre_prompt": "You are a helpful coding assistant. Respond concisely, but ensure all essential details are provided. Each of your statements must be unique.",
    "pre_prompt_suffix": "<|end_of_turn|>",
    "pre_prompt_prefix": "GPT4 System: "
  }
}

I have been further testing code generation, and I am impressed! It seems be almost on par with GPT4, and it can do things GPT4 cannot, like writing code for Google Apps Script. I have tried a few relatively complex tasks, in various languages (Powershell, Python, Javascript, Google Apps Script), covering various domain (Active Directory, Spotify, Mathematics). As far as I can tell, the code provided is correct!

Try the following example (make sure to load the OpenChat code prompt format I provided earlier); the result is impressive:

I need help writing code for google apps script. Could you please write a function that connects to spotify and sort a given playlist by order of popularity. Then another function to write back the sorted track to a given playlist. Make sure to handle pagination properly, in case the spotify playlist is large. Also added errors and exception handling. Include details of how to connect to spotify.

I think I found the key to avoid repetitions and long rambling answers, which this model has a tendency to do. Hopefully a further fine tune will reduce it. The key is to reduce creativity all the way down, and make the model deterministic. How do you do that?, you may ask. Easy, it is controlled by the following 3 inference parameters: temp, top_p, and top_k

With the following default settings I often get repetitions or additional rambling information:

    "top_k": 40,
    "top_p": 0.95,
    "temp": 0.8,

If I use the following values instead, to make the model deterministic, the problem seems to be gone:

    "top_k": 1,
    "top_p": 0.1,
    "temp": 0.1,

Please note that if you want to use the model for story writing, maybe you get better results by dialing up the creativity.

Here is my complete config file for LM Studio:

{
  "name": "OpenChat",
  "inference_params": {
    "top_k": 1,
    "top_p": 0.1,
    "temp": 0.1,
    "input_prefix": "GPT4 Correct User: ",
    "input_suffix": "<|end_of_turn|>GPT4 Correct Assistant: ",
    "antiprompt": [
      "GPT4",
      "<|end_of_turn|>",
      "[End of Turn]",
      "[]"
    ],
    "pre_prompt": "Below is an instruction that describes a task. Write a concise response that appropriately completes the request. Ensure all essential details are provided. Each of your statements must be unique.",
    "pre_prompt_suffix": "<|end_of_turn|>",
    "pre_prompt_prefix": "GPT4 System: "
  }
}

A few words about the above:

• I only include necessary options to avoid overwriting user settings when loading the model or switching prompt format. If you export a config file, please make sure you then edit it manually to clean it up.
• GPT Correct User/Assistant. The Correct keyword is important. It refers to the training data, where the answers were verified as correct. If you do not use it (eg: GPT4 User), it will still works, but it will five more weight to training data which was unverified (Human User was also used)
• GPT4 Sytem or just System are the 2 official recommended ways to prefix system messages. Either work.
• In my system message (pre_promt), I avoid any negative (eg: No repetitions). Remember this is just a language model: if it sees the word “repeat” (or similar), it will have a tendency to see this as an instruction to create repetitions! Instead I turned it around into a positive statement based on the word “unique”.
• Trailing spaces in the prefixes and suffixes are not critical, but ensure proper formatting.

​

As a bonus, here is my config for generating code, which according to my limited testing, this model seems to be surprisingly good at:

{
  "name": "OpenChat Code",
  "inference_params": {
    "top_k": 1,
    "top_p": 0.1,
    "temp": 0.1,
    "input_prefix": "Code User: ",
    "input_suffix": "<|end_of_turn|>Code Assistant: ",
    "antiprompt": [
      "GPT4",
      "<|end_of_turn|>",
      "[End of Turn]",
      "[]"
    ],
    "pre_prompt": "You are a helpful coding assistant. Respond concisely, but ensure all essential details are provided. Each of your statements must be unique.",
    "pre_prompt_suffix": "<|end_of_turn|>",
    "pre_prompt_prefix": "GPT4 System: "
  }
}

I noticed the same: in LM Studio I cannot enable Apple Metal (GPU), and I get the message: “Metal acceleration is not yet supported for this model architecture (‘starcoder’)”. However, according to Activity Monitor, it fully uses the GPU when it runs. And it is very fast!