tiiuae/Falcon3-1B-Instruct
Full fine-tune vs tiiuae/Falcon3-1B-Base. Fine-tune: 100% of the weights changed, across 18 of 18 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 37/37, MLP 54/54, Attention 72/72
15 derivatives checked
- bartowski/Falcon3-1B-Instruct-GGUF: Same weights, converted
- mradermacher/Falcon3-1B-Instruct-i1-GGUF: Quantization
- tiiuae/Falcon3-1B-Instruct-GGUF: Same weights, converted
- second-state/Falcon3-1B-Instruct-GGUF: Same weights, converted
- EasierAI/Falcon-3-1B: Quantization
- tiiuae/Falcon3-1B-Instruct-1.58bit: Quantization
- MaziyarPanahi/Falcon3-1B-Instruct-GGUF: Same weights, converted
- mradermacher/Falcon3-1B-Instruct-GGUF: Same weights, converted
- ChiKoi7/Falcon3-1B-Instruct-Heretic-GGUF: Abliterated
- lmstudio-community/Falcon3-1B-Instruct-GGUF: Quantization
- gaianet/Falcon3-1B-Instruct-GGUF: Same weights, converted
- tiiuae/Falcon3-1B-Instruct-GPTQ-Int4: Quantization
- tiiuae/Falcon3-1B-Instruct-GPTQ-Int8: Quantization
- ShivomH/Falcon3-1B-MentalHealth: LoRA-style
- ChiKoi7/Falcon3-1B-Instruct-Heretic: Abliterated
1 other repos hold byte-for-byte identical weight files.