jambroz/FNCARLplus-7b

TEXT GENERATIONConcurrency Cost:1Model Size:7BQuant:FP8Ctx Length:8kPublished:Apr 5, 2024License:apache-2.0Architecture:Transformer Open Weights Cold

jambroz/FNCARLplus-7b is a 7 billion parameter language model merged using the DARE TIES method, based on jambroz/sixtyoneeighty-7b. This model integrates capabilities from jambroz/FNCARL-7b, HuggingFaceH4/mistral-7b-anthropic, and mlabonne/UltraMerge-7B. It is designed to combine the strengths of its constituent models, offering a versatile foundation for various natural language processing tasks.

Loading preview...

Model Overview

jambroz/FNCARLplus-7b is a 7 billion parameter language model created through a sophisticated merge of several pre-trained models. It utilizes the DARE TIES merge method, a technique designed to combine the strengths of multiple models efficiently. The base model for this merge was jambroz/sixtyoneeighty-7b.

Key Merge Details

This model integrates components from three distinct sources:

  • jambroz/FNCARL-7b: Contributes to the model's overall linguistic understanding.
  • HuggingFaceH4/mistral-7b-anthropic: Likely enhances conversational and instruction-following capabilities, given its origin.
  • mlabonne/UltraMerge-7B: Suggests a focus on combining diverse model strengths.

The merge process involved specific density and weight parameters for each contributing model, indicating a fine-tuned approach to balancing their influences. The configuration also specified int8_mask: true, which can be relevant for quantization and efficiency.

Potential Use Cases

Given its merged nature, FNCARLplus-7b is suitable for a range of applications where a robust and versatile 7B parameter model is beneficial. It can be considered for tasks requiring general language understanding, generation, and potentially instruction-following, leveraging the combined knowledge of its constituent models.

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