eren23/ogno-monarch-jaskier-merge-7b-v2

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:4kPublished:Feb 20, 2024License:cc-by-nc-4.0Architecture:Transformer Open Weights Cold

The eren23/ogno-monarch-jaskier-merge-7b-v2 is a 7 billion parameter language model created by eren23, formed by merging several models including eren23/ogno-monarch-jaskier-merge-7b and mlabonne/AlphaMonarch-7B using the DARE TIES merging method. This model demonstrates strong general reasoning capabilities across various benchmarks, achieving an average score of 76.35 on the Open LLM Leaderboard. With a context length of 4096 tokens, it is suitable for a range of general-purpose language generation and understanding tasks.

Loading preview...

Overview

eren23/ogno-monarch-jaskier-merge-7b-v2 is a 7 billion parameter language model developed by eren23. It is a merge of multiple base models, specifically eren23/ogno-monarch-jaskier-merge-7b and mlabonne/AlphaMonarch-7B, utilizing the dare_ties merging method. The base model for this merge was eren23/dpo-binarized-NeutrixOmnibe-7B.

Key Capabilities

This model demonstrates robust performance across a variety of benchmarks, as evaluated on the Open LLM Leaderboard. Its key capabilities include:

  • Reasoning: Achieved 72.87 on the AI2 Reasoning Challenge (25-Shot).
  • Common Sense: Scored 89.15 on HellaSwag (10-Shot) and 84.29 on Winogrande (5-shot).
  • General Knowledge: Obtained 64.77 on MMLU (5-Shot) and 77.76 on TruthfulQA (0-shot).
  • Mathematical Reasoning: Performed with 69.29 on GSM8k (5-shot).

Good For

This model is well-suited for general-purpose language tasks requiring strong reasoning and common-sense understanding. Its balanced performance across diverse benchmarks makes it a versatile choice for applications such as:

  • Text generation and completion
  • Question answering
  • Summarization
  • Reasoning-based tasks

Finetunes
2 models