HuggingFaceH4/zephyr-7b-beta
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 16/65, MLP 96/96, Attention 128/128, lm_head 1/1
38 derivatives checked
- TheBloke/zephyr-7B-beta-GGUF: Quantization
- mradermacher/zephyr-7b-beta-GGUF: Quantization
- kukidevalml/LORAs: Different architecture
- QuantFactory/zephyr-7b-beta-GGUF: Quantization
- DevQuasar-8/HuggingFaceH4.zephyr-7b-beta-GGUF: Same weights, converted
- OPTML-Group/GradDiff-WMDP: Full fine-tune
- second-state/Zephyr-7B-Beta-GGUF: Quantization
- OPTML-Group/NPO-WMDP: Full fine-tune
- MaziyarPanahi/zephyr-7b-beta-GGUF: Quantization
- richardyoung/zephyr-7b-beta-abliterated: Abliterated
- OPTML-Group/NPO-SAM-WMDP: Full fine-tune
- BioMistral/BioMistral-7B-Zephyr-Beta-SLERP-GGUF: Full fine-tune
- tensorblock/zephyr-7b-beta-GGUF: Quantization
- TheBloke/zephyr-7B-beta-GPTQ: Quantization
- OPTML-Group/SimNPO-WMDP-zephyr-7b-beta: Full fine-tune
- TheBloke/zephyr-7B-beta-AWQ: Quantization
- wxzhang/dpo-selective-redteaming: Abliterated
- MexIvanov/zephyr-python-ru-merged: LoRA-style
- jambroz/sixtyoneeighty-7b: Full fine-tune
- grimjim/fireblossom-32K-7B: Full fine-tune
- ghost-x/ghost-7b-v0.9.0: LoRA-style
- CreitinGameplays/ConvAI-9b: Full fine-tune
- OPTML-Group/NPO-CR-WMDP: Full fine-tune
- BioMistral/BioMistral-7B-Zephyr-Beta-SLERP: Full fine-tune
- MuXodious/zephyr-7b-beta-absolute-heresy: Abliterated
- mohsinmubaraksk/Beast-Mixed: Full fine-tune
- grimjim/zephyr-beta-wizardLM-2-merge-7B: Full fine-tune
- lxyuan/AeolusBlend-7B-slerp: Full fine-tune
- bergr7f/ZephyrPaca-7B: Full fine-tune
- Dongwookss/futfut_by_zephyr7b: LoRA-style
15 other repos hold byte-for-byte identical weight files.