typhoon-ai/llama3.2-typhoon2-t1-3b-research-preview
Full fine-tune vs typhoon-ai/llama3.2-typhoon2-3b-instruct. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 44/57, MLP 84/84, Attention 112/112
4 derivatives checked
- mradermacher/llama-3.2-typhoon-t1-3b-research-preview-i1-GGUF: Full fine-tune
- mradermacher/llama3.2-typhoon2-t1-3b-research-preview-GGUF: Same weights, converted
- mradermacher/llama-3.2-typhoon-t1-3b-research-preview-GGUF: Full fine-tune
- typhoon-ai/llama3.2-typhoon2-t1-3b-research-preview-gguf: Same weights, converted