SylvanL/ChatTCM-7B-SFT
Full fine-tune vs SylvanL/ChatTCM-7B-Pretrain. Fine-tune: 92% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 0/1, Norms 57/57, MLP 84/84, Attention 196/196
2 derivatives checked
- mradermacher/ChatTCM-7B-SFT-i1-GGUF: Full fine-tune
- mradermacher/ChatTCM-7B-SFT-GGUF: Full fine-tune