tiiuae/Falcon-H1-3B-Instruct
Full fine-tune vs tiiuae/Falcon-H1-3B-Base. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: MLP 96/96, Norms 97/97, other 224/224, Attention 128/128
4 derivatives checked
- unsloth/Falcon-H1-3B-Instruct-GGUF: Same weights, converted
- mradermacher/Falcon-H1-3B-Instruct-i1-GGUF: Quantization
- DevQuasar/tiiuae.Falcon-H1-3B-Instruct-GGUF: Same weights, converted
- mradermacher/Falcon-H1-3B-Instruct-GGUF: Same weights, converted
1 other repos hold byte-for-byte identical weight files.