GestaltLabs/Ornstein-3.5-9B-V1.5
Partial fine-tune vs Qwen/Qwen3.5-9B-Base. Partial change: 82% of the weights changed (attention, lm_head, mlp, norms), in 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 0/5, MLP 96/211, Attention 178/336, other 0/1, Norms 98/221
4 derivatives checked
- mradermacher/Ornstein-3.5-9B-V1.5-GGUF: Same weights, converted
- mradermacher/Ornstein-3.5-9B-V1.5-i1-GGUF: Quantization
- GestaltLabs/Ornstein-3.5-9B-V1.5-GGUF: Quantization
- GestaltLabs/Ornstein-3.5-9B-V2: LoRA-style