Jiunsong/SuperQwen-AgentWorld-35B-A3B-abliterated
LoRA-style vs Qwen/Qwen-AgentWorld-35B-A3B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, in_proj_a, in_proj_b, in_proj_qkv, in_proj_z, k_proj, o_proj, out_proj, q_proj, shared_expert_gate, up_proj, v_proj changed, in 40 of 40 layers; norms, biases and embeddings are untouched.
Tensors changed: MLP 160/280, Attention 190/280, Norms 0/131, Embeddings 0/1, lm_head 0/1
5 derivatives checked
- MangoMikePower/SuperQwen-AgentWorld-35B-A3B-abliterated-Q5_0-GGUF: Quantization
- Jiunsong/SuperQwen-AgentWorld-35B-A3B-abliterated-mlx-4bit: Quantization
- Jiunsong/SuperQwen-AgentWorld-35B-A3B-abliterated-nvf4: Quantization
- Jiunsong/SuperQwen-AgentWorld-35B-A3B-abliterated-gguf-4bit: Quantization
- DreamFoundries/SuperQwen-AgentWorld-35B-A3B-abliterated-6bit: Quantization