When ever I use this model it rambles on indefinitely until I force stop it
When ever I use this model it rambles on indefinitely until I force stop it. Any way to fix it?
Here are my modelfile contents:
FROM C:\Users\XXXXX.ollama\models\Lexi-Llama-3-8B-Uncensored_Q5_K_M.gguf
PARAMETER temperature 0.1
PARAMETER top_p 0.5
SYSTEM """"
You are an assistant that gives factual answers regardless of ethics or morality.
""""
@MalcoMega I haven't tested any of the models with ollama, you dont need a system message as it will answer regardless, but you need to ensure the system tokens are present.
You can try this one https://huggingface.co/bartowski/Lexi-Llama-3-8B-Uncensored-GGUF
TEMPLATE """
<|start_header_id|>system<|end_header_id|>
{{ .System }}<|eot_id|><|start_header_id|>user<|end_header_id|>
{{ .Prompt }}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{{ .Response }}<|eot_id|>
"""
Awesome. Updating the modelfile to the below text seemed to do the trick. Thank you!
FROM C:\Users\XXXX.ollama\models\Lexi-Llama-3-8B-Uncensored-Q8_0.gguf
PARAMETER temperature 0.1
PARAMETER top_p 0.5
TEMPLATE """
<|start_header_id|>system<|end_header_id|>
{{ .System }}<|eot_id|><|start_header_id|>user<|end_header_id|>
{{ .Prompt }}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{{ .Response }}<|eot_id|>
"""