--- base_model: - gemma-2-2b-it/2 - Kukedlc/NeuralGemma2-2b-Spanish - Kukedlc/fusion_model_2 tags: - merge - mergekit - lazymergekit - gemma-2-2b-it/2 - Kukedlc/NeuralGemma2-2b-Spanish - Kukedlc/fusion_model_2 --- # NeuralGemma-2B-Spanish NeuralGemma-2B-Spanish is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing): * [gemma-2-2b-it/2](https://huggingface.co/gemma-2-2b-it/2) * [Kukedlc/NeuralGemma2-2b-Spanish](https://huggingface.co/Kukedlc/NeuralGemma2-2b-Spanish) * [Kukedlc/fusion_model_2](https://huggingface.co/Kukedlc/fusion_model_2) ## 🧩 Configuration ```yaml models: - model: gemma-2-2b/2 # No parameters necessary for base model - model: gemma-2-2b-it/2 parameters: density: 0.53 weight: 0.4 - model: Kukedlc/NeuralGemma2-2b-Spanish parameters: density: 0.44 weight: 0.2 - model: Kukedlc/fusion_model_2 parameters: density: 0.66 weight: 0.4 merge_method: dare_ties base_model: gemma-2-2b/2 parameters: int8_mask: true dtype: bfloat16 ``` ## 💻 Usage ```python !pip install -qU transformers accelerate from transformers import AutoTokenizer import transformers import torch model = "Kukedlc/NeuralGemma-2B-Spanish" messages = [{"role": "user", "content": "What is a large language model?"}] tokenizer = AutoTokenizer.from_pretrained(model) prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) pipeline = transformers.pipeline( "text-generation", model=model, torch_dtype=torch.float16, device_map="auto", ) outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95) print(outputs[0]["generated_text"]) ```