tiiuae/Falcon-H1-7B-Instruct
Full fine-tune vs tiiuae/Falcon-H1-7B-Base. Fine-tune: 100% of the weights changed, across 44 of 44 layers.
Tensors changed: MLP 132/132, Norms 133/133, other 308/308, Attention 176/176
10 derivatives checked
- tiiuae/Falcon-H1-7B-Instruct-GGUF: Same weights, converted
- unsloth/Falcon-H1-7B-Instruct-GGUF: Same weights, converted
- mradermacher/Falcon-H1-7B-Instruct-i1-GGUF: Quantization
- s3dev-ai/Falcon-H1-7B-Instruct-gguf: Same weights, converted
- mradermacher/Falcon-H1-7B-Instruct-GGUF: Same weights, converted
- tiiuae/Falcon-H1-7B-Instruct-GPTQ-Int4: Quantization
- DevQuasar/tiiuae.Falcon-H1-7B-Instruct-GGUF: Same weights, converted
- tiiuae/Falcon-H1-7B-Instruct-GPTQ-Int8: Quantization
- sgsystems/Falcon-H1-7B-FORGE-v2: Full fine-tune
- Vionex-digital/Hakeem-7B: Full fine-tune
1 other repos hold byte-for-byte identical weight files.