Lyon28 / caca-65M-untrained

🤖 caca-65M

Arsitektur Transformer Modern dengan Fitur Canggih

65,012,096 parameters • 65.01M • 77 layers • 2,048 tokens

📚 Documentation • 💻 Usage • ⚙️ Configuration • 🔬 Architecture

⚠️ PENTING: Model Belum Dilatih (Untrained)

Status Model:

• 🔴 Belum dilatih - Bobot masih random (Kaiming/Xavier init)
• 🟡 Untuk riset & eksperimen - Arsitektur sudah siap, tinggal train
• 🟢 Production-ready architecture - Teruji dan optimal

Widget di atas hanya menunjukkan format input yang diharapkan . Setelah model dilatih dengan dataset yang tepat, format yang sama akan menghasilkan output berkualitas tinggi.

🎯 Apa yang Bisa Dilakukan?

📋 Deskripsi

Caca adalah arsitektur Large Language Model (LLM) generasi terbaru yang menggabungkan berbagai teknik state-of-the-art dalam deep learning. Model ini dirancang dengan fokus pada efisiensi komputasi , skalabilitas , dan performa tinggi .

📖 Tentang Project Caca

Caca adalah eksperimen open-source Indonesian LLM yang dibuat dari nol secara individual dan bertahap. Bukan kompetitor siapa-siapa, cuma pengen eksplorasi apa yang bisa dilakukan dengan budget terbatas, passion unlimited, dan mindset collaborative.

Kalau berguna buat orang lain, alhamdulillah. Kalau enggak, ya tetap fun kok. Ini proyek eksplorasi, jadi kalau gagal ya bagian dari proses belajar. Kalau berhasil, itu bonus.

— Lyon , Creator

🌟 Mengapa Caca?

1. 🇮🇩 Fokus pada Bahasa Indonesia - Dirancang dengan mempertimbangkan karakteristik bahasa Indonesia
2. ⚡ Efisiensi Tinggi - GQA & Flash Attention untuk inferensi 3-5x lebih cepat
3. 💾 Memory Efficient - Hemat 50% memory untuk KV cache
4. 🔧 Modular & Extensible - Mudah dikustomisasi untuk berbagai use case
5. 🌐 Bilingual - Support optimal untuk Indonesia & English

🎯 Keunggulan vs Model Lain

🎯 Use Cases & Applications

✅ Cocok Untuk

❌ Tidak Cocok Untuk

📊 Spesifikasi Model

🎯 Core Features

🔥 Advanced Features

🎯 Mekanisme Attention

• ⚡ Flash Attention v2 - Algoritma IO-aware yang 3x lebih cepat dari attention standar
• 🔑 Grouped Query Attention (GQA) - 4 Query heads : 2 KV heads
  • Rasio kompresi: 2:1 (hemat ~50% memory KV cache)
• 🚀 xFormers Support - Fallback memory-efficient attention
• 🎯 PyTorch SDPA - Native scaled dot product attention

📍 Position Encodings

• 🔄 RoPE (Rotary Position Embeddings) - Base frequency θ=10,000
  • Generalisasi lebih baik untuk sequence panjang dibanding absolute PE

🎓 Optimisasi Training

• 💾 Gradient Checkpointing - Trade compute for memory (support model hingga 100B+ params)
• 🎯 Mixed Precision Training - Support FP16, BF16, dan TF32
• 📉 Dropout Regularization
  • Hidden dropout: 0.1
  • Attention dropout: 0.0
  • Residual dropout: 0.1

📦 Dukungan Quantization

• 4️⃣ 4-bit Quantization - NF4 & FP4 via bitsandbytes
  • Memory reduction: ~ 75% (4GB → 1GB)
  • Accuracy loss: <2% pada kebanyakan tasks
  • Support double quantization untuk kompresi maksimal
• 8️⃣ 8-bit Quantization - LLM.int8() dengan outlier handling
  • Memory reduction: ~ 50% (4GB → 2GB)
  • Accuracy loss: <1%
• 🔄 Dynamic Quantization - Runtime quantization tanpa calibration

🔬 Advanced Features

• 📊 Automatic Mixed Precision (AMP) - Dynamic loss scaling
• 🎯 Gradient Clipping - Stabilitas training dengan max norm clipping
• 📈 Learning Rate Scheduling - Support cosine, linear, warmup
• 💡 Smart Memory Management - Auto cache clearing & monitoring
• 🔍 Metrics Tracking - Real-time perplexity, loss, gradient norms
• 🛡️ NaN/Inf Detection - Automatic recovery dari numerical instability

💾 Kebutuhan Memory

Training Requirements

Inference Requirements

💡 Note : KV cache bertambah secara linear dengan panjang sequence. Untuk context 8K, kalikan nilai KV cache dengan 4.

Performance Estimates

📐 Struktur Arsitektur Lengkap

CacaForCausalLM (65.01M)
│
├─ Embedding: 16,000 × 256
│
├─ Transformer Layers (77x)
│  ├─ RMSNorm
│  ├─ Attention (GQA)
│  │  ├─ Q: 4 heads × 64 dim
│  │  ├─ KV: 2 heads × 64 dim
│  │  ├─ RoPE (θ=10,000)
│  │  └─ Flash Attention v2
│  ├─ Residual
│  ├─ RMSNorm
│  ├─ FFN (SwiGLU)
│  │  ├─ Gate: 256 → 704
│  │  ├─ Up: 256 → 704
│  │  └─ Down: 704 → 256
│  └─ Residual
│
├─ Final RMSNorm
└─ LM Head: 256 → 16,000

═══════════════════════════════════════════════════════════
📊 PARAMETER BREAKDOWN:
═══════════════════════════════════════════════════════════
Embeddings:                 4,096,000 (  6.3%)
Transformer Layers:        56,770,560 ( 87.3%)
  ├─ Attention:            15,138,816
  └─ FFN:                  41,631,744
Final Norm:                       256 (  0.0%)
───────────────────────────────────────────────────────────
TOTAL:                     65,012,096 (100.0%)
═══════════════════════════════════════════════════════════

📚 Dokumentasi

📦 Instalasi Dependencies

# Core dependencies (REQUIRED)
pip install torch>=2.0.0 transformers>=4.35.0 accelerate safetensors

# Optional: Untuk performa maksimal
pip install flash-attn --no-build-isolation  # Flash Attention 2 (3x speedup)
pip install xformers                          # Memory efficient attention
pip install bitsandbytes                      # 4/8-bit quantization

# Optional: Untuk monitoring & profiling
pip install tensorboard wandb               # Training monitoring
pip install gputil psutil                   # Resource monitoring

Compatibility Matrix:

Cara Penggunaan

1️⃣ Basic Loading

from transformers import AutoConfig, AutoModelForCausalLM, AutoTokenizer
import torch

# Load configuration
config = AutoConfig.from_pretrained(
    "Lyon28/caca-65M-untrained",
    trust_remote_code=True
)

# Load model (FP16 untuk efisiensi)
model = AutoModelForCausalLM.from_pretrained(
    "Lyon28/caca-65M-untrained",
    config=config,
    trust_remote_code=True,
    torch_dtype=torch.float16,
    device_map="auto"  # Automatic device placement
)

# Model ini UNTRAINED - butuh training dulu!
print(f"Model loaded: {model.num_parameters():,} parameters")
print("⚠️  Model ini belum dilatih dan belum bisa digunakan untuk inference")

2️⃣ Quantized Loading (4-bit/8-bit)

from transformers import AutoModelForCausalLM, BitsAndBytesConfig
import torch

# 4-bit quantization config
bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.bfloat16,
    bnb_4bit_use_double_quant=True
)

# Load model dengan quantization
model = AutoModelForCausalLM.from_pretrained(
    "Lyon28/caca-65M-untrained",
    trust_remote_code=True,
    quantization_config=bnb_config,
    device_map="auto"
)

print(f"Memory footprint: ~0.03GB (4-bit)")

3️⃣ Training Setup

from transformers import TrainingArguments, Trainer

# Training configuration
training_args = TrainingArguments(
    output_dir="./output",
    per_device_train_batch_size=1,
    gradient_accumulation_steps=16,
    learning_rate=2e-4,
    max_steps=10000,
    lr_scheduler_type="cosine",
    warmup_steps=500,
    logging_steps=10,
    save_steps=500,
    fp16=True,  # Mixed precision
    gradient_checkpointing=True,  # Memory efficient
)

# Initialize trainer
trainer = Trainer(
    model=model,
    args=training_args,
    train_dataset=train_dataset,
)

# Start training
trainer.train()

Advanced Usage

Gradient Checkpointing (Memory Efficient)

model.gradient_checkpointing_enable()
print("✅ Gradient checkpointing enabled - saves ~40% memory")

Custom Training Loop

from torch.optim import AdamW
from torch.cuda.amp import autocast, GradScaler

optimizer = AdamW(model.parameters(), lr=2e-4)
scaler = GradScaler()

for batch in dataloader:
    # Mixed precision forward
    with autocast(dtype=torch.bfloat16):
        outputs = model(**batch)
        loss = outputs.loss

    # Backward with gradient scaling
    scaler.scale(loss).backward()
    scaler.step(optimizer)
    scaler.update()
    optimizer.zero_grad()

Multi-GPU Training (DDP)

import torch.distributed as dist
from torch.nn.parallel import DistributedDataParallel

# Initialize process group
dist.init_process_group(backend="nccl")

# Wrap model
model = DistributedDataParallel(
    model,
    device_ids=[local_rank],
    find_unused_parameters=False
)

⚙️ Konfigurasi Detail

Full Configuration JSON

{
  "architectures": ["CacaForCausalLM"],
  "model_type": "caca",
  "vocab_size": 16000,
  "hidden_size": 256,
  "intermediate_size": 704,
  "num_hidden_layers": 77,
  "num_attention_heads": 4,
  "num_key_value_heads": 2,
  "head_dim": 64,
  "max_position_embeddings": 2048,
  "rope_theta": 10000,
  "rms_norm_eps": 1e-06,
  "use_cache": true,
  "use_qk_norm": true,
  "use_flash_attn": true,
  "attention_dropout": 0.0,
  "hidden_dropout": 0.1,
  "torch_dtype": "float16"
}

Custom Configuration

from transformers import AutoConfig

# Load dan modifikasi config
config = AutoConfig.from_pretrained("Lyon28/caca-65M-untrained")

# Custom modifications
config.max_position_embeddings = 16384  # Extend context
config.rope_scaling = {"type": "linear", "factor": 2.0}
config.use_flash_attn = True
config.hidden_dropout = 0.05

# Save custom config
config.save_pretrained("./custom_config")

🔬 Arsitektur

Layer Structure

Input Tokens
    ↓
Embedding Layer (16,000 → 256)
    ↓
┌─────────────────────────────────────┐
│ Decoder Block × 77                  │
│                                     │
│  ┌─ RMSNorm                        │
│  ├─ Multi-Head Attention (GQA)     │
│  │  - Flash Attention v2           │
│  │  - 4 Query heads, 2 KV heads       │
│  │  - RoPE position encoding       │
│  ├─ Residual Connection            │
│  │                                  │
│  ├─ RMSNorm                        │
│  ├─ Feed-Forward Network (SwiGLU)  │
│  │  - Gate: 256 → 704     │
│  │  - Up:   256 → 704     │
│  │  - Down: 704 → 256     │
│  └─ Residual Connection            │
│                                     │
└─────────────────────────────────────┘
    ↓
RMSNorm (Final)
    ↓
LM Head (256 → 16,000)
    ↓
Output Logits

Attention Mechanism (GQA)

Query:  [4 heads × 64 dim] = 256
Key:    [2 heads × 64 dim] = 128
Value:  [2 heads × 64 dim] = 128

Grouped Query Attention:
- Setiap 2 query heads berbagi 1 KV head
- Memory KV cache: 50% lebih kecil dari Multi-Head Attention
- Kualitas mendekati MHA, speed mendekati MQA

Feed-Forward Network (SwiGLU)

FFN(x) = (SiLU(xW_gate) ⊙ xW_up) W_down

Where:
- W_gate: 256 × 704
- W_up:   256 × 704
- W_down: 704 × 256
- SiLU(x) = x · sigmoid(x)
- ⊙ = element-wise multiplication

💬 Format Chat & Prompt Engineering

📝 Chat Template

Model mendukung format chat standar untuk conversational AI:

# Format chat template bawaan
chat_template = """
{% for message in messages %}
{% if message['role'] == 'system' %}
System: {{ message['content'] }}

{% elif message['role'] == 'user' %}
User: {{ message['content'] }}

{% elif message['role'] == 'assistant' %}
Assistant: {{ message['content'] }}

{% endif %}
{% endfor %}
{% if add_generation_prompt %}Assistant:{% endif %}
"""

# Contoh penggunaan
messages = [
    {"role": "system", "content": "Kamu adalah asisten AI yang membantu dan ramah."},
    {"role": "user", "content": "Jelaskan tentang fotosintesis"},
    {"role": "assistant", "content": "Fotosintesis adalah proses di mana tumbuhan mengubah cahaya matahari menjadi energi kimia..."},
    {"role": "user", "content": "Apa manfaatnya bagi manusia?"},
]

# Apply template
formatted = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)

print(formatted)
# Output:
# System: Kamu adalah asisten AI yang membantu dan ramah.
#
# User: Jelaskan tentang fotosintesis
# Assistant: Fotosintesis adalah proses di mana tumbuhan...
# User: Apa manfaatnya bagi manusia?
# Assistant:

🎯 Use Cases

Model ini dirancang untuk berbagai aplikasi NLP setelah melalui proses training:

Text Generation

• ✍️ Creative writing & storytelling
• 📰 Article generation
• 💬 Conversational AI
• 🔄 Text completion

Language Understanding

• 📊 Text classification
• 🏷️ Named Entity Recognition (NER)
• ❓ Question Answering
• 📝 Summarization

Code Generation

• 💻 Code completion
• 🐛 Bug fixing suggestions
• 📚 Documentation generation
• 🔄 Code translation

Multilingual Tasks

• 🌏 Translation (ID ↔ EN)
• 🗣️ Cross-lingual understanding
• 🌐 Multilingual classification

📈 Benchmark & Evaluation

⚠️ Model belum melalui evaluasi karena status untrained

Setelah training, model akan dievaluasi pada:

Indonesian Benchmarks

• IndoNLU : Comprehensive Indonesian NLU tasks
• IndoQA : Indonesian Question Answering
• IndoSum : Summarization
• IndoNER : Named Entity Recognition

Multilingual Benchmarks

• MMLU : Massive Multitask Language Understanding
• HellaSwag : Common sense reasoning
• ARC : Science QA
• TruthfulQA : Truthfulness evaluation

Generation Quality

• Perplexity : Language modeling quality
• BLEU/ROUGE : Translation & summarization
• Human Evaluation : Fluency, coherence, factuality

🛠️ Development & Training Tips

Optimal Batch Size

# Rule of thumb untuk 65.01M model
# GPU Memory → Batch size per device

if gpu_memory >= 80:  # A100 80GB
    batch_size = 246
    gradient_accumulation = 1
elif gpu_memory >= 40:  # A100 40GB
    batch_size = 123
    gradient_accumulation = 1
elif gpu_memory >= 24:  # RTX 3090/4090
    batch_size = 1
    gradient_accumulation = 1

# Effective batch size = batch_size × gradient_accumulation × num_gpus

Learning Rate Scheduling

# Recommended untuk 65.01M model
learning_rate = 0.0005  # Base LR
warmup_ratio = 0.05  # 5% of total steps
lr_scheduler = "cosine"  # atau "linear"

# Learning rate scaling rule:
# LR ∝ sqrt(batch_size)
# Untuk batch size 256: LR = 0.0005
# Untuk batch size 512: LR = 7.07e-04

Gradient Clipping

# Prevent gradient explosion
max_grad_norm = 1.0  # Clip at 1.0

# Monitor gradients
from torch.nn.utils import clip_grad_norm_

grad_norm = clip_grad_norm_(model.parameters(), max_grad_norm)
if grad_norm > 10.0:
    print(f"⚠️ High gradient norm: {grad_norm:.2f}")

Training Stability

# Tips untuk stable training:

1. **Warmup**: Mulai dengan LR rendah
2. **Gradient Checkpointing**: Kurangi memory footprint
3. **Mixed Precision**: Gunakan BF16 jika tersedia (lebih stable dari FP16)
4. **Batch Size**: Start small, increase gradually
5. **Monitor**: Track loss, perplexity, gradient norms

🔧 Troubleshooting

Out of Memory (OOM)

# Solusi OOM saat training:

✅ 1. Enable gradient checkpointing
model.gradient_checkpointing_enable()

✅ 2. Reduce batch size
per_device_train_batch_size = 1

✅ 3. Increase gradient accumulation
gradient_accumulation_steps = 32

✅ 4. Use quantization
load_in_8bit = True  # atau load_in_4bit

✅ 5. Reduce sequence length
max_length = 2048  # Start dengan ini

✅ 6. CPU offloading (jika perlu)
device_map = "auto"
offload_folder = "offload"

Slow Training

# Optimasi kecepatan training:

✅ 1. Flash Attention
config.use_flash_attn = True  # 2-3x speedup

✅ 2. Compile model (PyTorch 2.0+)
model = torch.compile(model, mode="reduce-overhead")

✅ 3. DataLoader optimization
dataloader = DataLoader(
    dataset,
    batch_size=batch_size,
    num_workers=4,  # Parallel data loading
    pin_memory=True,  # Faster GPU transfer
    prefetch_factor=2
)

✅ 4. Mixed precision
use_fp16 = True  # atau bf16

✅ 5. Optimize communication (multi-GPU)
find_unused_parameters = False
gradient_as_bucket_view = True

NaN Loss

# Jika loss menjadi NaN:

✅ 1. Reduce learning rate
learning_rate = learning_rate * 0.1

✅ 2. Check gradient norms
clip_grad_norm_(model.parameters(), 1.0)

✅ 3. Use BF16 instead of FP16
torch_dtype = torch.bfloat16  # Lebih stable

✅ 4. Add epsilon to RMSNorm
rms_norm_eps = 1e-5  # Increase jika perlu

✅ 5. Check data
# Pastikan tidak ada inf/nan di dataset
assert not torch.isnan(input_ids).any()
assert not torch.isinf(attention_mask).any()

🚫 Prohibited Uses

📜 License & Citation

📄 License

Full license text : Apache-2.0

📖 Citation

Jika Anda menggunakan model ini dalam penelitian, mohon sitasi:

@misc{cacacaca65m,
  author = {Lyon},
  title = {Caca-caca-65M: Modern Transformer Architecture with Grouped Query Attention},
  year = {2026},
  publisher = {Hugging Face},
  journal = {Hugging Face Model Hub},
  howpublished = {\url{https://huggingface.co/Lyon28/caca-65M-untrained}},
  note = {Untrained model with 65,012,096 parameters}
}

APA Style:

Lyon. (2026). Caca-caca-65M: Modern Transformer Architecture with Grouped
Query Attention [Untrained model]. Hugging Face.
https://huggingface.co/Lyon28/caca-65M-untrained

MLA Style:

Lyon. "Caca-caca-65M: Modern Transformer Architecture with Grouped Query Attention."
Hugging Face, 2026, huggingface.co/Lyon28/caca-65M-untrained.

🙏 Acknowledgments

Model ini berdiri di pundak para raksasa! Terima kasih kepada:

📄 License

Model ini dirilis di bawah Apache License 2.0 .

Ketentuan Penggunaan:

• ✅ Bebas digunakan untuk keperluan komersial dan non-komersial
• ✅ Modifikasi diperbolehkan
• ✅ Distribusi diperbolehkan dengan attribution
• ⚠️ No Warranty - model disediakan "as is"
• 📝 Attribution Required - sertakan copyright notice

Lihat LICENSE untuk detail lengkap.

🤝 Contributing

Kami sangat terbuka untuk kontribusi! Berikut cara Anda bisa berkontribusi:

Training & Fine-tuning

• 🎓 Train model ini dengan dataset Anda
• 📊 Share benchmark results
• 🔬 Experiment dengan hyperparameters

Code & Architecture

• 🐛 Report bugs atau issues
• 💡 Suggest improvements
• 🔧 Submit pull requests

Documentation

• 📚 Improve documentation
• 🌐 Add translations
• ✍️ Write tutorials & guides

Dataset & Evaluation

• 📝 Contribute training data
• 🧪 Create evaluation benchmarks
• 🎯 Share fine-tuned versions

👥 Team & Acknowledgments

Core Team

• LyonPoy - Architecture design & implementation

Special Thanks

• 🤗 Hugging Face - Infrastructure & community
• ⚡ FlashAttention Team - Efficient attention implementation
• 🧠 Anthropic, Google, Meta - Research inspirations

Community

Terima kasih kepada komunitas open-source yang telah berkontribusi pada:

• Transformers library
• PyTorch framework
• Datasets & evaluation tools

📞 Contact & Support

Community

• 💬 Discussions - Ask questions
• 🐛 Issues - Report bugs
• 📧 Email : [email protected]

🌟 Star History

💝 Dibuat dengan ❤️ untuk Komunitas AI Indonesia

Terima kasih telah menggunakan Caca!

Jika model ini berguna, jangan lupa ⭐ repository kami!

📊 Model Statistics

🎨 Daily Inspiration

📈 Quick Stats

🌟 "Dari nol, untuk semua" 🌟

_{Last updated: january 2026}