sail/Sailor2-8B-Chat
Full fine-tune vs sail/Sailor2-8B. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 56/65, MLP 96/96, Attention 224/224, lm_head 1/1
8 derivatives checked
- bartowski/Sailor2-8B-Chat-GGUF: Same weights, converted
- MaziyarPanahi/Sailor2-8B-Chat-GGUF: Same weights, converted
- mradermacher/Sailor2-8B-Chat-GGUF: Same weights, converted
- DevQuasar-5/sail.Sailor2-8B-Chat-GGUF: Quantization
- BlossomsAI/BloomVN-8B-chat: Full fine-tune
- BlossomsAI/Sailor2-8B-Chat-Uncensored: Abliterated
- JosephNguyen/Sailor2-8B-Chat-finetuned: LoRA-style
- real-jiakai/Sailor2-8B-Chat-Jiakai: LoRA-style