Instructions to use oswaldoludwig/kappatune-lora-tinyllama-agnews with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use oswaldoludwig/kappatune-lora-tinyllama-agnews with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("TinyLlama/TinyLlama-1.1B-Chat-v1.0")
model = PeftModel.from_pretrained(base_model, "oswaldoludwig/kappatune-lora-tinyllama-agnews")

Transformers

How to use oswaldoludwig/kappatune-lora-tinyllama-agnews with Transformers:

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

pipe = pipeline("text-generation", model="oswaldoludwig/kappatune-lora-tinyllama-agnews")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("oswaldoludwig/kappatune-lora-tinyllama-agnews", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use oswaldoludwig/kappatune-lora-tinyllama-agnews with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "oswaldoludwig/kappatune-lora-tinyllama-agnews"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "oswaldoludwig/kappatune-lora-tinyllama-agnews",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/oswaldoludwig/kappatune-lora-tinyllama-agnews

SGLang

How to use oswaldoludwig/kappatune-lora-tinyllama-agnews 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 "oswaldoludwig/kappatune-lora-tinyllama-agnews" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "oswaldoludwig/kappatune-lora-tinyllama-agnews",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "oswaldoludwig/kappatune-lora-tinyllama-agnews" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "oswaldoludwig/kappatune-lora-tinyllama-agnews",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use oswaldoludwig/kappatune-lora-tinyllama-agnews with Docker Model Runner:
```
docker model run hf.co/oswaldoludwig/kappatune-lora-tinyllama-agnews
```

oswaldoludwig commited on Mar 13

Commit

addac3c

verified ·

1 Parent(s): a661295

Update README.md

Browse files

Files changed (1) hide show

README.md +3 -3

README.md CHANGED Viewed

@@ -31,9 +31,9 @@ This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-Chat-v1.0](https
 ## Model description
 This model is a LoRA (Low-Rank Adaptation) adapter applied to TinyLlama-1.1B. Unlike standard LoRA, which targets manually specified module types (e.g., all `q_proj` or `v_proj` layers), this adapter was trained using the **KappaTune PEFT Integration**.
-Before training, the KappaTune algorithm performed a Singular Value Decomposition (SVD) on all candidate weight matrices to calculate their **Condition Number (\( \kappa \))**.
-* **High-\( \kappa \)** tensors (highly specialized, anisotropic weights containing pre-trained knowledge) were frozen.
-* **Low-\( \kappa \)** tensors (numerically stable, general-purpose weights acting as a "raw marble block") were selected for LoRA adaptation.
 This targeted approach ensures the model learns the new domain efficiently while preserving its foundational conversational capabilities.

 ## Model description
 This model is a LoRA (Low-Rank Adaptation) adapter applied to TinyLlama-1.1B. Unlike standard LoRA, which targets manually specified module types (e.g., all `q_proj` or `v_proj` layers), this adapter was trained using the **KappaTune PEFT Integration**.
+Before training, the KappaTune algorithm performed a Singular Value Decomposition (SVD) on all candidate weight matrices to calculate their **Condition Number (kappa)**.
+* **High-kappa** tensors (highly specialized, anisotropic weights containing pre-trained knowledge) were frozen.
+* **Low-kappa** tensors (numerically stable, general-purpose weights of higher output entropy acting as a "raw marble block") were selected for LoRA adaptation.
 This targeted approach ensures the model learns the new domain efficiently while preserving its foundational conversational capabilities.