neopolita/jessi-v0.4-falcon3-7b-instruct
Full fine-tune vs tiiuae/Falcon3-7B-Instruct. Fine-tune: 99% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 0/57, MLP 84/84, Attention 84/112, lm_head 1/1
3 derivatives checked
- mradermacher/jessi-v0.4-falcon3-7b-instruct-i1-GGUF: Quantization
- mradermacher/jessi-v0.4-falcon3-7b-instruct-GGUF: Same weights, converted
- suayptalha/Falcon3-Jessi-v0.4-7B-Slerp: Full fine-tune