Kadabra/Gemma4-e2b-CPT
Kadabra/Gemma4-e2b-CPT is a 5.1 billion parameter Gemma4-based causal language model developed by Kadabra. This model was fine-tuned using Unsloth and Huggingface's TRL library, enabling 2x faster training. It is designed for general language generation tasks, leveraging its efficient training methodology.
Loading preview...
Model Overview
Kadabra/Gemma4-e2b-CPT is a 5.1 billion parameter language model developed by Kadabra. It is based on the Gemma4 architecture and was fine-tuned from the unsloth/gemma-4-e2b-unsloth-bnb-4bit model. A key differentiator of this model is its training methodology, which utilized Unsloth and Huggingface's TRL library to achieve a 2x speedup in the fine-tuning process.
Key Characteristics
- Architecture: Gemma4-based causal language model.
- Parameter Count: 5.1 billion parameters.
- Context Length: Supports a context length of 32768 tokens.
- Training Efficiency: Fine-tuned with Unsloth, resulting in significantly faster training times.
- License: Released under the Apache-2.0 license.
Use Cases
This model is suitable for a variety of general language generation and understanding tasks, benefiting from its efficient fine-tuning process. Its substantial context window makes it capable of handling longer inputs and generating more coherent, extended outputs.