saucam/Pyrhea-72B
LoRA-style (fewer layers) vs abacusai/Smaug-72B-v0.1. saucam/Pyrhea-72B matches part of abacusai/Smaug-72B-v0.1 (723 of its 1043 tensors; the rest of abacusai/Smaug-72B-v0.1 has no counterpart). On the shared part: 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 80 of 80 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/161, MLP 240/240, Attention 320/320, lm_head 0/1
2 derivatives checked
- mradermacher/Pyrhea-72B-i1-GGUF: Quantization
- mradermacher/Pyrhea-72B-GGUF: Quantization