Instructions to use norallm/normistral-7b-scratch with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use norallm/normistral-7b-scratch with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="norallm/normistral-7b-scratch")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("norallm/normistral-7b-scratch")
model = AutoModelForCausalLM.from_pretrained("norallm/normistral-7b-scratch")

llama-cpp-python

How to use norallm/normistral-7b-scratch with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="norallm/normistral-7b-scratch",
	filename="normistral-7b-scratch.Q4_K_M.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Inference
Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use norallm/normistral-7b-scratch with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf norallm/normistral-7b-scratch:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf norallm/normistral-7b-scratch:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf norallm/normistral-7b-scratch:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf norallm/normistral-7b-scratch:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf norallm/normistral-7b-scratch:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf norallm/normistral-7b-scratch:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf norallm/normistral-7b-scratch:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf norallm/normistral-7b-scratch:Q4_K_M

Use Docker

docker model run hf.co/norallm/normistral-7b-scratch:Q4_K_M

LM Studio
Jan

vLLM

How to use norallm/normistral-7b-scratch with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "norallm/normistral-7b-scratch"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "norallm/normistral-7b-scratch",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/norallm/normistral-7b-scratch:Q4_K_M

SGLang

How to use norallm/normistral-7b-scratch with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "norallm/normistral-7b-scratch" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "norallm/normistral-7b-scratch",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "norallm/normistral-7b-scratch" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "norallm/normistral-7b-scratch",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use norallm/normistral-7b-scratch with Ollama:
```
ollama run hf.co/norallm/normistral-7b-scratch:Q4_K_M
```

Unsloth Studio new

How to use norallm/normistral-7b-scratch with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for norallm/normistral-7b-scratch to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for norallm/normistral-7b-scratch to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for norallm/normistral-7b-scratch to start chatting

Docker Model Runner
How to use norallm/normistral-7b-scratch with Docker Model Runner:
```
docker model run hf.co/norallm/normistral-7b-scratch:Q4_K_M
```

Lemonade

How to use norallm/normistral-7b-scratch with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull norallm/normistral-7b-scratch:Q4_K_M

Run and chat with the model

lemonade run user.normistral-7b-scratch-Q4_K_M

List all available models

lemonade list

Unable to finetune using axolotl

by henriklied - opened Feb 7, 2024

Discussion

henriklied

Feb 7, 2024

I tried fine tuning the model using axolotl, but it seems like your special_tokens.json has some issues and looks quite different from the main Mistral one as well as Nb's.

I ended up copying the base Mistral specials_tokens_map.json and setting tokenizer_type: PreTrainedTokenizerFast per your tokenizer_config.json, but that just crashed with TypeError: MistralForCausalLM.forward() got an unexpected keyword argument 'token_type_ids'.

I ran into the same issues on the -warm model.

Any tips would be greatly appreciated.

davda54

Norwegian Large Language Models org Feb 7, 2024

Hi, our models use a Norwegian tokenizer, which is different from the mostly English tokenizer of Mistral. But the special_tokens_map.json file is in the standard HuggingFace format,

It would be really helpful if you could send us a minimal reproducible example of this bug or at least a detailed error trace.

henriklied

Feb 8, 2024

Hi again,

Sorry, I meant the tokenizer_config.json file, not the special_tokens_map.json. It seems like axolotl was unable to process it.
I copied Mistrals tokenizer_config.json and changed tokenizer_class to PreTrainedTokenizerFast, as well as purged all my dataset caches, and now I'm able to fine tune the model. ☺️

I'm using the alpaca format for fine tuning it right now. The default system prompt for alpaca fine tuning is "Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request". Should I use this, or would you recommend swapping it out for something else when instruction fine tuning? The rest of my instruction, user and system messages are all in Norwegian.

davda54

Norwegian Large Language Models org Feb 8, 2024

Hi,

That's great news! Thanks for finding a way around this issue. However, note that the tokenizers are completely different (SentencePiece vs. HF tokenizers), so copying the full config from Mistral might introduce some unwanted artifacts. Do you by any chance remember what item(s) was axolotl missing from the config file? Adding only this item(s) would be a safer workaround, I would also fix it directly in this repository.

Norwegian system prompt will probably work better for these models :)

henriklied

Feb 9, 2024

The error message was quite cryptic, with sentencepiece/init.py line 310, in LoadFromFilespitting out"TypeError: not a string"`.
The fine tuning ended up with an eval loss of 1.077 on 3 epochs, fine tuned on 120k articles where the output was title generation. Feels like it performs better than my testing with GPT-J, so that's fun. I'll do another run with a Norwegian system prompt as well. :-)

buzzcraft

Feb 9, 2024

Want to share your tokenizer_config.json setup @henriklied ?

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment