Elvenmaid 12B Stock

🚀 ElvenMaid-12B-Stock Model

This is a ChatML model created by merging pre - trained language models, offering enhanced capabilities for various language - related tasks.

🚀 Quick Start

This is a merged pre - trained language model created using mergekit.

✨ Features

• Multi - language support: Supports both English (en) and Japanese (ja).
• Based on ChatML: Follows the ChatML standard for better interaction in chat scenarios.

📚 Documentation

📦 Installation

Since the library_name is transformers, you can install it using the following command:

pip install transformers

💻 Usage Examples

Although no specific code examples are provided in the original document, a general usage example using the transformers library might look like this:

from transformers import AutoModelForCausalLM, AutoTokenizer

# Load the model and tokenizer
model_name = "yamatazen/ElvenMaid-12B-Stock"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

# Prepare input text
input_text = "Your input text here"
input_ids = tokenizer(input_text, return_tensors='pt').input_ids

# Generate output
output = model.generate(input_ids)
output_text = tokenizer.decode(output[0], skip_special_tokens=True)
print(output_text)

🔧 Technical Details

Merge Details

Merge Method

This model was merged using the Model Stock merge method, with yamatazen/ElvenMaid-12B as the base model.

Models Merged

The following models were included in the merge:

• yamatazen/ElvenMaid-12B-v2
• yamatazen/ElvenMaid-12B-v3

Configuration

The following YAML configuration was used to produce this model:

base_model: yamatazen/ElvenMaid-12B
models:
  - model: yamatazen/ElvenMaid-12B-v2
  - model: yamatazen/ElvenMaid-12B-v3
merge_method: model_stock
dtype: bfloat16
parameters:
  normalize: true

📄 Information Table