Upload folder using huggingface_hub
Browse files- README.md +53 -134
- pytorch_model.pt +3 -0
README.md
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
---
|
| 2 |
-
license:
|
| 3 |
language:
|
| 4 |
- ko
|
| 5 |
tags:
|
|
@@ -9,164 +9,83 @@ tags:
|
|
| 9 |
- attention
|
| 10 |
- griffin
|
| 11 |
- language-model
|
| 12 |
-
datasets:
|
| 13 |
-
- wikipedia
|
| 14 |
-
pipeline_tag: text-generation
|
| 15 |
---
|
| 16 |
|
| 17 |
-
# HybriKo
|
| 18 |
|
| 19 |
-
Griffin
|
| 20 |
|
| 21 |
-
##
|
| 22 |
|
| 23 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
|
| 25 |
-
##
|
| 26 |
-
- **타입**: Hybrid RNN-Attention (2:1 비율)
|
| 27 |
-
- **파라미터**: 117.8M
|
| 28 |
-
- **Hidden Dimension**: 768
|
| 29 |
-
- **레이어**: 12개 (8 RNN + 4 Attention)
|
| 30 |
-
- **Attention Heads**: 12개 (GQA, 3 KV heads)
|
| 31 |
-
- **Vocabulary**: 32,000 (SentencePiece Unigram)
|
| 32 |
-
- **최대 시퀀스 길이**: 512
|
| 33 |
|
| 34 |
-
|
| 35 |
-
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
Layer 4-5: GriffinBlock
|
| 39 |
-
Layer 6: AttentionBlock
|
| 40 |
-
... (패턴 반복)
|
| 41 |
-
```
|
| 42 |
-
|
| 43 |
-
### 아키텍처 다이어그램
|
| 44 |
-
|
| 45 |
-

|
| 46 |
-
|
| 47 |
-
## 학습
|
| 48 |
|
| 49 |
-
|
| 50 |
-
- **학습 스텝**: 1,000
|
| 51 |
-
- **배치 사이즈**: 16
|
| 52 |
-
- **Learning Rate**: 3e-4
|
| 53 |
-
- **하드웨어**: A100
|
| 54 |
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
|
| 60 |
-
|
|
|
|
|
|
|
|
|
|
| 61 |
|
| 62 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 63 |
|
| 64 |
-
|
| 65 |
|
| 66 |
```python
|
| 67 |
-
# 의존성 설치
|
| 68 |
-
!pip install transformers sentencepiece -q
|
| 69 |
-
|
| 70 |
import torch
|
| 71 |
-
|
| 72 |
-
from
|
| 73 |
-
|
| 74 |
-
# GPU 자동 감지
|
| 75 |
-
device = "cuda" if torch.cuda.is_available() else "cpu"
|
| 76 |
-
print(f"🖥️ 사용 디바이스: {device}")
|
| 77 |
-
|
| 78 |
-
# 모델 코드 다운로드
|
| 79 |
-
config_path = hf_hub_download("Yaongi/HybriKo-117M", "configuration_hybridko.py")
|
| 80 |
-
model_path = hf_hub_download("Yaongi/HybriKo-117M", "modeling_hybridko.py")
|
| 81 |
|
| 82 |
-
|
| 83 |
-
|
| 84 |
-
|
| 85 |
-
from configuration_hybridko import HybriKoConfig
|
| 86 |
-
from modeling_hybridko import HybriKoModel
|
| 87 |
-
|
| 88 |
-
# 1. 모델 생성
|
| 89 |
-
config = HybriKoConfig()
|
| 90 |
model = HybriKoModel(config)
|
|
|
|
| 91 |
|
| 92 |
-
#
|
| 93 |
-
|
| 94 |
-
checkpoints = sorted([f for f in files if f.startswith("checkpoint_step_") and f.endswith(".pt")])
|
| 95 |
-
latest_checkpoint = checkpoints[-1] if checkpoints else "checkpoint_step_1000.pt"
|
| 96 |
-
print(f"📦 체크포인트 로드: {latest_checkpoint}")
|
| 97 |
-
|
| 98 |
-
checkpoint_path = hf_hub_download("Yaongi/HybriKo-117M", latest_checkpoint)
|
| 99 |
-
checkpoint = torch.load(checkpoint_path, map_location=device)
|
| 100 |
-
model.load_state_dict(checkpoint["model_state_dict"])
|
| 101 |
-
model = model.to(device)
|
| 102 |
-
model.eval()
|
| 103 |
-
|
| 104 |
-
# 3. 토크나이저 로드
|
| 105 |
-
tokenizer_path = hf_hub_download("Yaongi/HybriKo-117M", "HybriKo_tok.model")
|
| 106 |
-
sp = spm.SentencePieceProcessor()
|
| 107 |
-
sp.Load(tokenizer_path)
|
| 108 |
-
|
| 109 |
-
# 4. 텍스트 생성
|
| 110 |
-
prompt = "한국의 수도는"
|
| 111 |
-
input_ids = torch.tensor([[2] + sp.EncodeAsIds(prompt)]).to(device)
|
| 112 |
-
output = model.generate(input_ids, max_new_tokens=50, temperature=0.8)
|
| 113 |
-
print(sp.DecodeIds(output[0].tolist()))
|
| 114 |
-
```
|
| 115 |
|
| 116 |
-
#
|
| 117 |
-
|
| 118 |
-
|
| 119 |
-
|
| 120 |
-
|
| 121 |
-
for prompt in prompts:
|
| 122 |
-
input_ids = torch.tensor([[2] + sp.EncodeAsIds(prompt)]).to(device) # 👈 .to(device) 추가
|
| 123 |
-
output = model.generate(input_ids, max_new_tokens=30, temperature=0.8, top_k=50)
|
| 124 |
-
generated = sp.DecodeIds(output[0].tolist())
|
| 125 |
-
print(f"📝 {prompt}")
|
| 126 |
-
print(f" → {generated}")
|
| 127 |
-
print("-" * 50)
|
| 128 |
```
|
| 129 |
|
| 130 |
-
##
|
| 131 |
-
|
| 132 |
-
| 파라미터 | 설명 | 권장 값 |
|
| 133 |
-
|----------|------|---------|
|
| 134 |
-
| `temperature` | 랜덤성 (낮을수록 결정적) | 0.7 - 1.0 |
|
| 135 |
-
| `top_k` | 상위 K개 토큰만 샘플링 | 50 |
|
| 136 |
-
| `top_p` | Nucleus 샘플링 임계값 | 0.9 |
|
| 137 |
-
| `max_new_tokens` | 생성할 토큰 수 | 30 - 100 |
|
| 138 |
-
|
| 139 |
-
## 제한 사항
|
| 140 |
-
|
| 141 |
-
⚠️ **이 모델은 1,000 스텝만 학습된 개념 증명(Proof-of-Concept) 모델입니다.**
|
| 142 |
-
|
| 143 |
-
생성된 텍스트는 초기 학습 패턴(숫자/연도 많이 출력)을 보입니다. 더 나은 품질을 위해:
|
| 144 |
-
- 10,000+ 스텝 학습
|
| 145 |
-
- 1B+ 파라미터로 스케일업
|
| 146 |
-
- 특정 태스크에 파인튜닝
|
| 147 |
-
|
| 148 |
-
## 레포지토리 파일
|
| 149 |
|
| 150 |
-
|
| 151 |
-
|
| 152 |
-
|
| 153 |
-
|
| 154 |
-
| `HybriKo_tok.vocab` | 토크나이저 어휘 |
|
| 155 |
-
| `config.json` | HuggingFace 설정 |
|
| 156 |
-
| `configuration_hybridko.py` | Config 클래스 |
|
| 157 |
-
| `modeling_hybridko.py` | 모델 아키텍처 |
|
| 158 |
|
| 159 |
-
##
|
| 160 |
|
| 161 |
```bibtex
|
| 162 |
-
@misc{
|
| 163 |
-
title={HybriKo:
|
| 164 |
-
|
| 165 |
-
|
| 166 |
-
url={https://huggingface.co/Yaongi/HybriKo-117M}
|
| 167 |
}
|
| 168 |
```
|
| 169 |
|
| 170 |
-
##
|
| 171 |
|
| 172 |
-
|
|
|
|
| 1 |
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
language:
|
| 4 |
- ko
|
| 5 |
tags:
|
|
|
|
| 9 |
- attention
|
| 10 |
- griffin
|
| 11 |
- language-model
|
|
|
|
|
|
|
|
|
|
| 12 |
---
|
| 13 |
|
| 14 |
+
# HybriKo: Korean Hybrid Language Model
|
| 15 |
|
| 16 |
+
A Griffin-inspired hybrid architecture combining RNN and Attention mechanisms for Korean language modeling.
|
| 17 |
|
| 18 |
+
## Model Details
|
| 19 |
|
| 20 |
+
- **Parameters**: 117.8M
|
| 21 |
+
- **Architecture**: 2:1 RNN-to-Attention ratio (Griffin-inspired)
|
| 22 |
+
- **Context Length**: 1024 tokens
|
| 23 |
+
- **Vocab Size**: 32,000 (SentencePiece)
|
| 24 |
+
- **Training Data**: Korean Wikipedia
|
| 25 |
|
| 26 |
+
## Training Results (Exp3)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
|
| 28 |
+
| Phase | Steps | Loss | PPL |
|
| 29 |
+
|-------|-------|------|-----|
|
| 30 |
+
| Phase 1 | 0-10K | 1.80 | ~6.0 |
|
| 31 |
+
| Phase 2 | 10K-30K | 1.60 | ~4.95 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
|
| 33 |
+
## Architecture
|
|
|
|
|
|
|
|
|
|
|
|
|
| 34 |
|
| 35 |
+
```
|
| 36 |
+
HybriKo (117.8M params)
|
| 37 |
+
├── Embedding (32000 → 768)
|
| 38 |
+
├── Layers (12x)
|
| 39 |
+
│ ├── Layer 1,2: GriffinBlock (RNN)
|
| 40 |
+
│ ├── Layer 3: AttentionBlock
|
| 41 |
+
│ └── (pattern repeats)
|
| 42 |
+
└── LM Head (weight-tied)
|
| 43 |
+
```
|
| 44 |
|
| 45 |
+
Key features:
|
| 46 |
+
- **RGLRU**: Real-Gated Linear Recurrent Unit
|
| 47 |
+
- **GQA**: Grouped Query Attention (1:4 KV reduction)
|
| 48 |
+
- **Flash Attention 2**: Optimized attention computation
|
| 49 |
+
- **GeGLU**: Gated activation in FFN
|
| 50 |
|
| 51 |
+
## Usage
|
| 52 |
|
| 53 |
```python
|
|
|
|
|
|
|
|
|
|
| 54 |
import torch
|
| 55 |
+
from hybridko.model import HybriKoModel, HybriKoConfig
|
| 56 |
+
from hybridko.data import load_tokenizer
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 57 |
|
| 58 |
+
# Load model
|
| 59 |
+
config = HybriKoConfig.from_yaml("config.yaml")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
model = HybriKoModel(config)
|
| 61 |
+
model.load_state_dict(torch.load("pytorch_model.pt"))
|
| 62 |
|
| 63 |
+
# Load tokenizer
|
| 64 |
+
tokenizer = load_tokenizer("HybriKo_tok.model")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 65 |
|
| 66 |
+
# Generate
|
| 67 |
+
from hybridko.inference import generate_with_cache
|
| 68 |
+
output = generate_with_cache(model, tokenizer, "한국의 수도는", max_tokens=50)
|
| 69 |
+
print(output)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 70 |
```
|
| 71 |
|
| 72 |
+
## Files
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
|
| 74 |
+
- `pytorch_model.pt`: Model weights (450MB)
|
| 75 |
+
- `config.yaml`: Model configuration
|
| 76 |
+
- `HybriKo_tok.model`: SentencePiece tokenizer
|
| 77 |
+
- `HybriKo_tok.vocab`: Tokenizer vocabulary
|
|
|
|
|
|
|
|
|
|
|
|
|
| 78 |
|
| 79 |
+
## Citation
|
| 80 |
|
| 81 |
```bibtex
|
| 82 |
+
@misc{hybridko2024,
|
| 83 |
+
title={HybriKo: Korean Hybrid Language Model},
|
| 84 |
+
year={2024},
|
| 85 |
+
url={https://huggingface.co/gyunggyung/HybriKo-117M}
|
|
|
|
| 86 |
}
|
| 87 |
```
|
| 88 |
|
| 89 |
+
## License
|
| 90 |
|
| 91 |
+
Apache 2.0
|
pytorch_model.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e02a596f9dc1a993cd1aa0a65022a5d4ec95409620be3c43f2829432e93b5077
|
| 3 |
+
size 471349067
|