Felladrin/TinyMistral-248M-Chat-v4
Full fine-tune vs Locutusque/TinyMistral-248M. Fine-tune: 100% of the weights changed, across 12 of 12 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 25/25, MLP 36/36, Attention 48/48
11 derivatives checked
- afrideva/TinyMistral-248M-Alpaca-GGUF: Full fine-tune
- afrideva/TinyMistral-248M-SFT-v4-GGUF: Full fine-tune
- afrideva/TinyMistral-248M-Evol-Instruct-GGUF: Full fine-tune
- MaziyarPanahi/TinyMistral-248M-SFT-v4-GGUF: Full fine-tune
- afrideva/TinyMistral-248M-SFT-v3-GGUF: Full fine-tune
- mradermacher/TinyMistral-248M-Chat-v4-GGUF: Same weights, converted
- M4-ai/TinyMistral-6x248M: MoE built from dense models
- Felladrin/gguf-TinyMistral-248M-SFT-v4: Full fine-tune
- Felladrin/gguf-TinyMistral-248M-Chat-v2: Full fine-tune
- balnazzar/Smolly-248M: Full fine-tune
- LeroyDyer/SpydazWebAI_MiniAI_248m: Full fine-tune