HuggingFaceH4/zephyr-7b-alpha
Full fine-tune vs mistralai/Mistral-7B-v0.1. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 17/65, MLP 96/96, Attention 128/128, lm_head 1/1
8 derivatives checked
- mradermacher/zephyr-7b-alpha-GGUF: Quantization
- TheBloke/zephyr-7B-alpha-GGUF: Quantization
- QuantFactory/zephyr-7b-alpha-GGUF: Quantization
- second-state/Zephyr-7B-Alpha-GGUF: Quantization
- TheBloke/zephyr-7B-alpha-AWQ: Quantization
- MaziyarPanahi/zephyr-7b-alpha-GGUF: Quantization
- sethuiyer/SynthIQ-7b: Full fine-tune
- Sakalti/Magro-7b-v1.1: Full fine-tune
6 other repos hold byte-for-byte identical weight files.