Instructions to use MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("MagistrTheOne/KNK-VF-Lab-38B")
model = PeftModel.from_pretrained(base_model, "MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA")

Transformers

How to use MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA with Transformers:

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

pipe = pipeline("text-generation", model="MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA

SGLang

How to use MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA 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 "MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA" \
    --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": "MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA",
		"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 "MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA" \
        --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": "MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA with Docker Model Runner:
```
docker model run hf.co/MagistrTheOne/KUROTAMA-KNK-VF-Lab38B-SFT-LoRA
```

KUROTAMA-KNK-VF-Lab38B-SFT-LoRA

File size: 1,578 Bytes

5d8edc7

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 29,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 112.91146087646484,
      "learning_rate": 8.43120818934367e-05,
      "loss": 178.4693359375,
      "step": 10
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 69.01692962646484,
      "learning_rate": 3.019601169804216e-05,
      "loss": 143.72413330078126,
      "step": 20
    },
    {
      "epoch": 1.0,
      "eval_loss": 8.519869804382324,
      "eval_runtime": 12.6349,
      "eval_samples_per_second": 3.166,
      "eval_steps_per_second": 3.166,
      "step": 29
    },
    {
      "epoch": 1.0,
      "step": 29,
      "total_flos": 2.9686113589837824e+16,
      "train_loss": 151.82884058459052,
      "train_runtime": 802.9271,
      "train_samples_per_second": 0.573,
      "train_steps_per_second": 0.036
    }
  ],
  "logging_steps": 10,
  "max_steps": 29,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.9686113589837824e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}