adonlee/LLaMA_2_13B_SFT_v0

TEXT GENERATIONConcurrency Cost:1Model Size:13BQuant:FP8Ctx Length:4kLicense:apache-2.0Architecture:Transformer Open Weights Cold

adonlee/LLaMA_2_13B_SFT_v0 is a 13 billion parameter language model based on the LLaMA 2 architecture, fine-tuned for general instruction following. This model demonstrates competitive performance across various benchmarks, including ARC, HellaSwag, and MMLU, making it suitable for a range of natural language understanding and generation tasks. Its supervised fine-tuning (SFT) aims to enhance its ability to respond accurately and coherently to user prompts.

Loading preview...

Overview

adonlee/LLaMA_2_13B_SFT_v0 is a 13 billion parameter language model built upon the LLaMA 2 architecture, specifically enhanced through supervised fine-tuning (SFT). This model is designed to follow instructions effectively, providing coherent and relevant responses across a variety of general-purpose natural language tasks.

Key Capabilities & Performance

This model has been evaluated on the Open LLM Leaderboard, showcasing its performance across several key benchmarks:

  • Average Score: 50.97
  • ARC (25-shot): 62.03
  • HellaSwag (10-shot): 83.8
  • MMLU (5-shot): 58.39
  • TruthfulQA (0-shot): 49.92
  • Winogrande (5-shot): 77.27
  • GSM8K (5-shot): 12.43
  • DROP (3-shot): 12.96

These scores indicate its proficiency in common sense reasoning, reading comprehension, and general knowledge tasks, with a context length of 4096 tokens.

Good For

  • General instruction-following applications.
  • Tasks requiring common sense reasoning and factual recall.
  • Prototyping and development where a LLaMA 2-based 13B model with SFT is desired.

Popular Sampler Settings

Top 3 parameter combinations used by Featherless users for this model. Click a tab to see each config.

temperature
top_p
top_k
frequency_penalty
presence_penalty
repetition_penalty
min_p