SicariusSicariiStuff/Impish_QWEN_14B-1M
LoRA-style vs Qwen/Qwen2.5-14B-Instruct-1M. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 48 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/97, MLP 144/144, Attention 192/336, lm_head 0/1
6 derivatives checked
- SicariusSicariiStuff/Impish_QWEN_14B-1M_GGUF: Quantization
- mradermacher/Impish_QWEN_14B-1M-i1-GGUF: Quantization
- mradermacher/Impish_QWEN_14B-1M-GGUF: Quantization
- SicariusSicariiStuff/Impish_QWEN_14B-1M_iMatrix: Quantization
- SicariusSicariiStuff/Impish_QWEN_14B-1M_GGUF_HA: Quantization
- win10/miscii-14b-1M-0128: Full fine-tune