Instructions to use werty1248/EXAONE-3.5-32B-LIMO-Ko-e4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use werty1248/EXAONE-3.5-32B-LIMO-Ko-e4 with Transformers:

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

pipe = pipeline("text-generation", model="werty1248/EXAONE-3.5-32B-LIMO-Ko-e4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("werty1248/EXAONE-3.5-32B-LIMO-Ko-e4")
model = AutoModelForCausalLM.from_pretrained("werty1248/EXAONE-3.5-32B-LIMO-Ko-e4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use werty1248/EXAONE-3.5-32B-LIMO-Ko-e4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "werty1248/EXAONE-3.5-32B-LIMO-Ko-e4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "werty1248/EXAONE-3.5-32B-LIMO-Ko-e4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/werty1248/EXAONE-3.5-32B-LIMO-Ko-e4

SGLang

How to use werty1248/EXAONE-3.5-32B-LIMO-Ko-e4 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 "werty1248/EXAONE-3.5-32B-LIMO-Ko-e4" \
    --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": "werty1248/EXAONE-3.5-32B-LIMO-Ko-e4",
		"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 "werty1248/EXAONE-3.5-32B-LIMO-Ko-e4" \
        --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": "werty1248/EXAONE-3.5-32B-LIMO-Ko-e4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use werty1248/EXAONE-3.5-32B-LIMO-Ko-e4 with Docker Model Runner:
```
docker model run hf.co/werty1248/EXAONE-3.5-32B-LIMO-Ko-e4
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

데이터 셋

LIMO

GAIR/LIMO (영어, 원본)

LIMO 한국어 번역

exp-models/GAIR-LIMO-KOREAN (한국어 번역)
junnei/ko-limo (한국어 번역)

특이사항

원래 LIMO에서는 15 epoch 학습을 수행함
영어1+한국어2 데이터 셋을 섞은 후 5 epoch 학습시켜 원래 학습 방법과 유사한 횟수만큼, 그러나 약간의 변형이 있도록 학습시키려고 함
그러나 정성 평가에서 4 epoch 시점의 checkpoint가 가장 성능이 좋아 보였음

Training Details

4xH200 SXM, 13.5 Hours

Axolotl config

base_model: beomi/EXAONE-3.5-32B-Instruct-Llamafied
model_type: AutoModelForCausalLM
tokenizer_config: beomi/EXAONE-3.5-32B-Instruct-Llamafied
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  - path: werty1248/kk_oo_llliiimmmooo
    field_messages: conversations
    type: chat_template
    chat_template: tokenizer_default

dataset_prepared_path: ./data_preparation
output_dir: /workspace/data

hf_use_auth_token: true

sequence_len: 32768
sample_packing: false
pad_to_sequence_len: true

plugins:
  - axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_layer_norm: true
liger_glu_activation: true
liger_fused_linear_cross_entropy: true

wandb_project:
#wandb_entity:
#wandb_watch:
wandb_name:
#wandb_log_model:

gradient_accumulation_steps: 2
micro_batch_size: 1
num_epochs: 5
optimizer: paged_adamw_8bit
lr_scheduler: cosine
learning_rate: 5.0e-6

train_on_inputs: false
group_by_length: false
bf16: auto
fp16: 
tf32: false

gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_ratio: 0.05
eval_table_size:

save_total_limit: 2

deepspeed: ./deepspeed_configs/zero3_bf16.json

special_tokens:
  pad_token: "[|endofturn|]"

Downloads last month: 3

Safetensors

Model size

32B params

Tensor type

BF16

Model tree for werty1248/EXAONE-3.5-32B-LIMO-Ko-e4

Base model

LGAI-EXAONE/EXAONE-3.5-32B-Instruct

Finetuned

(6)

this model

Quantizations

1 model

werty1248
/

EXAONE-3.5-32B-LIMO-Ko-e4