hwkwon/S-SOLAR-10.7B-v1.4

TEXT GENERATIONConcurrent Unit Cost:1Model Size:10.7BQuant:FP8Context Size:4kLicense:cc-by-nc-4.0Architecture:Transformer Open Weights Featherless Exclusive Cold

S-SOLAR-10.7B-v1.4 is a 10.7 billion parameter language model developed by hwkwon, fine-tuned from chihoonlee10/T3Q-ko-solar-dpo-v1.0 using DeepSpeed. This model is designed for general language generation tasks, building upon its base model's capabilities. Its fine-tuning process aims to enhance performance for specific applications, making it suitable for non-commercial use under the cc-by-nc-4.0 license.

Loading preview...

Model Overview

hwkwon/S-SOLAR-10.7B-v1.4 is a 10.7 billion parameter language model that has been fine-tuned from the existing chihoonlee10/T3Q-ko-solar-dpo-v1.0 model. The fine-tuning process utilized DeepSpeed, a deep learning optimization library, suggesting an emphasis on efficient training and potentially improved performance characteristics over its base model. While specific details on the training data are currently TBA (To Be Announced), the model is designed to handle general language tasks.

Key Characteristics

  • Base Model: Fine-tuned from chihoonlee10/T3Q-ko-solar-dpo-v1.0.
  • Training Method: Utilizes DeepSpeed for fine-tuning, indicating potential performance enhancements.
  • Prompt Format: Employs a simple ### User: User query input\n\n### Assistant: template for interaction.

Licensing and Usage

This model is distributed under the cc-by-nc-4.0 license, which permits sharing and adaptation for non-commercial purposes only. Users should adhere to these terms when integrating the model into their applications or research.

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