| --- |
| library_name: transformers |
| tags: |
| - ai |
| - chatty |
| - hoopoe2.4 |
| - conversational |
| license: apache-2.0 |
| language: |
| - he |
| - en |
| base_model: |
| - Raziel1234/Duchifat-2 |
| pipeline_tag: text-generation |
| --- |
| |
| # Duchifat-2.4-Instruct (136M) ๐ฆ |
|
|
| **Duchifat-2.4-Instruct** represents a significant evolution in the Duchifat series. This version (2.4) is a specialized, instruction-tuned model that has been refined through a massive training pipeline to achieve state-of-the-art performance for its size (136M parameters). |
|
|
| ## ๐ Whatโs New in Version 2.4? |
|
|
| Version 2.4 is not just a minor update; it's a complete refinement of the model's behavior and alignment: |
|
|
| - **Advanced Token Density:** v2.4 has been pushed to a total of **3.27 Billion tokens**, ensuring the model has reached peak saturation for its 136M architecture. |
| - **Structural Alignment:** Unlike previous iterations, 2.4 is natively aligned to the `<|instruction|>` and `<|assistant|>` tokens. The model now treats these as fundamental structural boundaries. |
| - **Hard-Coded EOS Logic:** We have fixed the termination issues from earlier versions. v2.4 is specifically trained to predict and emit the `<|eos|>` token at the precise end of every instruction and response block, ensuring clean and reliable chat sessions. |
| - **Improved Hebrew Fluency:** v2.4 leverages the DictaLM-3.0-24B tokenizer logic more effectively, resulting in a more natural "flow" of the Hebrew language without the stuttering found in smaller models. |
|
|
| ## ๐ Technical Highlights |
|
|
| - **Model Version:** 2.4 (Instruct) |
| - **Parameter Count:** 136M |
| - **Training Scale:** 3.27B Tokens (Mixed C4 Hebrew/English) |
| - **Architecture:** Optimized Transformer with RoPE and RMSNorm. |
| - **Inference Speed:** Ultra-low latency, ideal for real-time bilingual applications. |
|
|
| ## ๐ป Implementation (v2.4) |
|
|
| To utilize the improved logic of v2.4, ensure you use `trust_remote_code=True` and follow the mandatory format. |
|
|
| ```python |
| import torch |
| from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer |
| |
| # ืชืืงืื ืืืืืช ื-Instruct (ื-r ืืคื ื ื-u) |
| model_id = "razielAI/Hoopoe-2.4-Instruct" |
| |
| print(f"ืืืขื ืืช ืืืืื ืืฆืืืืจื {model_id}... ื ื ืืืืชืื.") |
| |
| try: |
| # ืืขืื ืช ืืืืงื ืืืืจ |
| tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True) |
| |
| # ืืขืื ืช ืืืืื |
| model = AutoModelForCausalLM.from_pretrained( |
| model_id, |
| torch_dtype=torch.bfloat16, |
| device_map="auto", |
| trust_remote_code=True |
| ) |
| |
| # ืกื ืืจืื ืืืื ื-Vocab |
| if model.get_input_embeddings().weight.shape[0] != len(tokenizer): |
| model.resize_token_embeddings(len(tokenizer)) |
| |
| streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=False) |
| |
| def run_chat(): |
| print(f"\n--- {model_id} Chat Ready ---") |
| model.eval() |
| while True: |
| user_input = input("\n๐ค ืืฉืชืืฉ: ") |
| if user_input.lower() in ["exit", "quit", "ืืฆืืื", "ืืื"]: |
| break |
| |
| prompt = f"<|instruction|>{user_input}<|eos|><|assistant|>" |
| inputs = tokenizer(prompt, return_tensors="pt", add_special_tokens=False).to(model.device) |
| |
| print("๐ค Hoopoe: ", end="") |
| with torch.no_grad(): |
| model.generate( |
| input_ids=inputs["input_ids"], |
| attention_mask=inputs["attention_mask"], |
| max_new_tokens=512, |
| temperature=0.7, |
| do_sample=True, |
| pad_token_id=tokenizer.eos_token_id, |
| eos_token_id=tokenizer.eos_token_id, |
| repetition_penalty=1.15, |
| streamer=streamer |
| ) |
| print() |
| |
| if __name__ == "__main__": |
| run_chat() |
| |
| except Exception as e: |
| print(f"\nืฉืืืื ืืืขืื ื: {e}") |
| print("\nืขืฆื: ืื ืก ืืืฃ ืืืืื ื-Hugging Face ืืชืืืื ืฉืฉื ืืืฉืชืืฉ ืืืืืื ืืชืืืื ืืืืืง ืื.") |
| ``` |
