passthepizza's picture
Update README.md
7ebf0ce verified
|
raw
history blame
1.48 kB
metadata
license: mit
language:
  - en
base_model:
  - meta-llama/Llama-3.1-70B-Instruct
tags:
  - axolotl

🎭 Cakrawala-70B

Where Worlds Converge and Adventures Begin!

🌟 What's Special About This Model?

Cakrawala-70B is a fine-tuned variant of the Llama-3.1-70B-Instruct model, specifically optimised for generating rich roleplaying conversations and character interactions. The model has been trained to excel at producing detailed, contextually appropriate character dialogues with rich descriptions of physical actions, expressions, and emotional states while maintaining consistent character voices and perspectives throughout extended interactions.

πŸ§ͺ The Secret Sauce

Training Diet:

  • Fed with 5,867 conversation pairs
  • Each conversation is a minimum 12-13 turns long
  • Focused heavily details like facial expressions, environmental descriptions, and character reactions that are focused a lot on keeping the model in character.

Tech Wizardry:

  • Trained on the mighty Llama-3.1-70B-Instruct
  • Fine-tuned using QLoRA
  • Trained over 3 epochs

Training Parameters

  • Gradient Accumulation Steps: 16
  • Micro Batch Size: 4
  • Learning Rate: 0.0003
  • Optimizer: AdamW
  • Scheduler: Cosine
  • Mixed Precision: BF16 & FP16 with TF32 support

πŸ”§ Under the Hood

  • Trained on 8 x H100 NVL GPUs

🎬 License & Credits

  • Licensed under MIT
  • Based on meta-llama/Llama-3.1-70B-Instruct

Built with ❀️ for roleplayers, by roleplayers