66 lines
1.4 KiB
Markdown
66 lines
1.4 KiB
Markdown
---
|
|
base_model: unsloth/phi-3.5-mini-instruct-bnb-4bit
|
|
language:
|
|
- en
|
|
license: apache-2.0
|
|
tags:
|
|
- text-generation-inference
|
|
- transformers
|
|
- unsloth
|
|
- phi
|
|
- gguf
|
|
datasets:
|
|
- mlabonne/FineTome-100k
|
|
- microsoft/orca-math-word-problems-200k
|
|
- m-a-p/CodeFeedback-Filtered-Instruction
|
|
- cognitivecomputations/dolphin-coder
|
|
- PawanKrd/math-gpt-4o-200k
|
|
- V3N0M/Jenna-50K-Alpaca-Uncensored
|
|
---
|
|
|
|
# Phi-3.5-mini-instruct-uncensored
|
|
|
|
- **Developed by:** Carsen Klock
|
|
- **License:** apache-2.0
|
|
- **Finetuned from model :** unsloth/phi-3.5-mini-instruct-bnb-4bit
|
|
|
|
This Phi3.5 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
|
|
|
|
Trained on 1 x 4080 SUPER over 10500 Epochs as a test. This is for test purposes only.
|
|
|
|
GGUFs are included in this repository for inference
|
|
|
|
|
|
Running in transformers
|
|
```py
|
|
# Use a pipeline as a high-level helper
|
|
from transformers import pipeline
|
|
|
|
messages = [
|
|
{"role": "user", "content": "Who are you?"},
|
|
]
|
|
pipe = pipeline("text-generation", model="carsenk/phi3.5_mini_exp_825_uncensored")
|
|
pipe(messages)
|
|
|
|
print(pipe(messages))
|
|
```
|
|
|
|
|
|
Running in llama.cpp (Use GGUF)
|
|
```py
|
|
from llama_cpp import Llama
|
|
|
|
llm = Llama.from_pretrained(
|
|
repo_id="carsenk/phi3.5_mini_exp_825_uncensored",
|
|
filename="unsloth.BF16.gguf",
|
|
)
|
|
|
|
llm.create_chat_completion(
|
|
messages = [
|
|
{
|
|
"role": "user",
|
|
"content": "What is the capital of France?"
|
|
}
|
|
]
|
|
)
|
|
``` |