openchat/openchat-3.5-0106
Full fine-tune vs mistralai/Mistral-7B-v0.1. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Norms 65/65, MLP 96/96, Attention 128/128
21 derivatives checked
- TheBloke/openchat-3.5-0106-GGUF: Quantization
- mradermacher/openchat-3.5-0106-i1-GGUF: Quantization
- mradermacher/openchat-3.5-0106-GGUF: Same weights, converted
- mayacinka/chatty-djinn-14B: Full fine-tune (fewer layers)
- second-state/OpenChat-3.5-0106-GGUF: Quantization
- MaziyarPanahi/openchat-3.5-0106-GGUF: Quantization
- zeio/openchat-3.5-0106-GGUF: Quantization
- newsletter/openchat-3.5-0106-Q6_K-GGUF: Quantization
- LoneStriker/Newton-7B-GGUF: Full fine-tune
- sethuiyer/Chikuma_10.7B: Full fine-tune (fewer layers)
- jieliu/Storm-7B: LoRA-style
- Pretergeek/OpenChat-3.5-0106_8.99B_40Layers-Appended: Same weights, repacked
- Pretergeek/OpenChat-3.5-0106_8.11B_36Layers-Appended: Same weights, repacked
- somosnlp/NoticIA-7B: LoRA-style
- Weyaxi/Newton-7B: Full fine-tune
- Pretergeek/OpenChat-3.5-0106_10.7B_48Layers-Appended: Same weights, repacked (fewer layers)
- andysalerno/openchat-nectar-0.1: LoRA-style
- Pretergeek/OpenChat-3.5-0106_9.86B_44Layers-Appended: Same weights, repacked
- andysalerno/openchat-nectar-0.5: LoRA-style
- Iker/ClickbaitFighter-7B: Full fine-tune
- mergekit-community/mergekit-slerp-euzaldk: Full fine-tune
8 other repos hold byte-for-byte identical weight files.