Hikari07jp/MiMo-V2.6-Distill-Qwen-9B-Ablitrated
LoRA-style vs XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, up_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Embeddings 0/4, MLP 40/208, Attention 0/332, Norms 0/215
9 derivatives checked
- mradermacher/MiMo-V2.6-Distill-Qwen-9B-ablit-i1-GGUF: Quantization
- mradermacher/MiMo-V2.6-Distill-Qwen-9B-Ablitrated-i1-GGUF: Quantization
- mradermacher/MiMo-V2.6-Distill-Qwen-9B-Ablitrated-GGUF: Same weights, converted
- mradermacher/MiMo-V2.6-Distill-Qwen-9B-ablit-GGUF: Same weights, converted
- xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Ablitrated-oQ4e: Quantization
- xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Ablitrated-oQ3e: Quantization
- xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Ablitrated-oQ5e: Quantization
- xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Ablitrated-Text-oQ4e: Quantization
- xunlinkx/MiMo-V2.6-Distill-Qwen-9B-Ablitrated-Text-oQ3e: Quantization