|
--- |
|
base_model: daryl149/llama-2-7b-chat-hf |
|
datasets: |
|
- Hypersniper/philosophy_dialogue |
|
--- |
|
|
|
# Overview |
|
Fine-tuned [Llama-2 7B](https://huggingface.co/daryl149/llama-2-7b-chat-hf) with a philosopher conversation dataset (originally from [Hypersniper/philosophy_dialogue](https://huggingface.co/datasets/Hypersniper/philosophy_dialogue)). |
|
Used QLoRA for fine-tuning. Trained for one epoch on a 40GB GPU (NVIDIA A100) instance. |
|
|
|
The version here is the fp16 HuggingFace model. |
|
|
|
# Prompt style |
|
The model was trained with the following prompt style: |
|
``` |
|
<s>[INST] <<SYS>> |
|
{{ system_prompt }} |
|
<</SYS>> |
|
|
|
{{ user_message }} [/INST] |
|
{{ response }} </s> |
|
``` |