Instructions to use BUT-FIT/csmpt7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use BUT-FIT/csmpt7b with Transformers:

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

pipe = pipeline("text-generation", model="BUT-FIT/csmpt7b", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("BUT-FIT/csmpt7b", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("BUT-FIT/csmpt7b", trust_remote_code=True)

llama-cpp-python

How to use BUT-FIT/csmpt7b with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="BUT-FIT/csmpt7b",
	filename="BUT-FIT_csmpt7b-6.7B-BF16.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use BUT-FIT/csmpt7b with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf BUT-FIT/csmpt7b:BF16
# Run inference directly in the terminal:
llama-cli -hf BUT-FIT/csmpt7b:BF16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf BUT-FIT/csmpt7b:BF16
# Run inference directly in the terminal:
llama-cli -hf BUT-FIT/csmpt7b:BF16

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 BUT-FIT/csmpt7b:BF16
# Run inference directly in the terminal:
./llama-cli -hf BUT-FIT/csmpt7b:BF16

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 BUT-FIT/csmpt7b:BF16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf BUT-FIT/csmpt7b:BF16

Use Docker

docker model run hf.co/BUT-FIT/csmpt7b:BF16

LM Studio
Jan

vLLM

How to use BUT-FIT/csmpt7b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "BUT-FIT/csmpt7b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BUT-FIT/csmpt7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/BUT-FIT/csmpt7b:BF16

SGLang

How to use BUT-FIT/csmpt7b 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 "BUT-FIT/csmpt7b" \
    --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": "BUT-FIT/csmpt7b",
		"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 "BUT-FIT/csmpt7b" \
        --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": "BUT-FIT/csmpt7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use BUT-FIT/csmpt7b with Ollama:
```
ollama run hf.co/BUT-FIT/csmpt7b:BF16
```

Unsloth Studio

How to use BUT-FIT/csmpt7b 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 BUT-FIT/csmpt7b 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 BUT-FIT/csmpt7b to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for BUT-FIT/csmpt7b to start chatting

Docker Model Runner
How to use BUT-FIT/csmpt7b with Docker Model Runner:
```
docker model run hf.co/BUT-FIT/csmpt7b:BF16
```

Lemonade

How to use BUT-FIT/csmpt7b with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull BUT-FIT/csmpt7b:BF16

Run and chat with the model

lemonade run user.csmpt7b-BF16

List all available models

lemonade list

mfajcik commited on Dec 16, 2024

Commit

7574425

verified ·

1 Parent(s): 17f6699

Update README.md

Browse files

Files changed (1) hide show

README.md +4 -1

README.md CHANGED Viewed

@@ -70,6 +70,10 @@ To transfer knowledge from English model to Czech, we developed a simple method
 <img src="figures/tllama_test.png" width="900"/>
 Figure 4: Test perplexity over the course of training for vocabulary swap (swapping 1.7K tokens) method on TinyLLAMA. Our method (green curve) vs TinyLLAMA training from scratch (blue curve).
 The vocabulary swap was done the same way as our [Czech-GPT-2](https://huggingface.co/BUT-FIT/Czech-GPT-2-XL-133k) model (check it out for comprehensive description.)
 For CSMPT7b, we managed to align 4,177 english tokens with corresponding czech tokens.
@@ -97,7 +101,6 @@ Not mentioned hyperparameters were kept the same as for MPT.
 | scheduler_steps            | 170,000       |                                                                                              |
 | scheduler_alpha            | 0.1           | So LR on last step is 0.1*(vanilla LR)                                                       |
 # Usage
 ## How to Setup Environment
 ```bash

 <img src="figures/tllama_test.png" width="900"/>
 Figure 4: Test perplexity over the course of training for vocabulary swap (swapping 1.7K tokens) method on TinyLLAMA. Our method (green curve) vs TinyLLAMA training from scratch (blue curve).
+We also verify that finetuning from English to Czech is beneficial for MPT-7B model, compared from training a new model, at least on the first 10K steps. The training also seems to be more stable (notice yellow spike around 10k steps).
+Figure 5: Test cross-entropy over the course of training on CSMPT7B (yellow-red). Comparison with TinyLLAMA (blue-green). Our method (red&green curve) vs TinyLLAMA training from scratch (yellow&blue curve).
+<img src="figures/csmpt_tllama_test.png" width="900"/>
 The vocabulary swap was done the same way as our [Czech-GPT-2](https://huggingface.co/BUT-FIT/Czech-GPT-2-XL-133k) model (check it out for comprehensive description.)
 For CSMPT7b, we managed to align 4,177 english tokens with corresponding czech tokens.
 | scheduler_steps            | 170,000       |                                                                                              |
 | scheduler_alpha            | 0.1           | So LR on last step is 0.1*(vanilla LR)                                                       |
 # Usage
 ## How to Setup Environment
 ```bash