HuggingFaceH4/zephyr-7b-gemma-sft-v0.1
Full fine-tune vs google/gemma-7b. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 51/57, MLP 84/84, Attention 112/112
6 derivatives checked
- MaziyarPanahi/zephyr-7b-gemma-sft-v0.1-GGUF: Quantization
- bartowski/DiscoPOP-zephyr-7b-gemma-GGUF: Full fine-tune
- QuantFactory/DiscoPOP-zephyr-7b-gemma-GGUF: Quantization
- LoneStriker/zephyr-7b-gemma-v0.1-GGUF: Quantization
- HuggingFaceH4/zephyr-7b-gemma-v0.1: Full fine-tune
- SakanaAI/DiscoPOP-zephyr-7b-gemma: Full fine-tune