tensopolis/lamarckvergence-14b-tensopolis-v1
LoRA-style vs suayptalha/Lamarckvergence-14B. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of gate_proj changed, in 12 of 48 layers; norms, biases and embeddings are untouched.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 0/97, MLP 12/144, Attention 0/336
2 derivatives checked
- mradermacher/lamarckvergence-14b-tensopolis-v1-i1-GGUF: Quantization
- mradermacher/lamarckvergence-14b-tensopolis-v1-GGUF: Quantization