# Qwen3 8M Model with Falcon-H1-0.5B-Instruct Tokenizer

## Model Description
This is an 8M parameter Qwen3 model architecture combined with the Falcon-H1-0.5B-Instruct tokenizer (32K vocabulary).

- **Architecture**: Qwen3 (Grouped Query Attention, RMS Normalization, Q/K Normalization, RoPE)
- **Tokenizer**: Falcon-H1-0.5B-Instruct (32K vocab)
- **Parameters**: 2,183,552
- **Precision**: BF16
- **Format**: SafeTensors
- **Vocabulary Size**: 32768

## Configuration
- vocab_size: 32768
- hidden_size: 64
- num_attention_heads: 4
- num_key_value_heads: 2
- num_hidden_layers: 2
- intermediate_size: 160
- head_dim: 16
- max_position_embeddings: 4096

## Special Tokens
- BOS: <|begin_of_text|> (id: 17)
- EOS: <|end_of_text|> (id: 11)
- PAD: <|pad|> (id: 0)

## Usage
```python
from transformers import Qwen3ForCausalLM, AutoTokenizer

model = Qwen3ForCausalLM.from_pretrained("./workspace/qwen3-8m-falcon-tokenizer")
tokenizer = AutoTokenizer.from_pretrained("./workspace/qwen3-8m-falcon-tokenizer")

# Generate text
inputs = tokenizer("Hello, world!", return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=50)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

# Batch processing (start small)
texts = ["Hello", "How are you", "Good morning"]
inputs = tokenizer(texts, return_tensors="pt", padding=True, truncation=True)
with torch.no_grad():
    outputs = model.generate(**inputs, max_new_tokens=20)
```

## Important Notes
- Model uses Qwen3 architecture with Falcon tokenizer (32K vocabulary)
- All token IDs must be < 32768 to avoid CUDA errors
- Start with small batch sizes (1-4) and gradually increase
- Use proper padding to prevent dimension mismatches
- Model initialized with random weights - requires fine-tuning
- Compatible with Qwen3 APIs but uses Falcon vocabulary