mattshumer/Hermes-2-Pro-11B
Partial fine-tune (fewer layers) vs NousResearch/Hermes-2-Pro-Mistral-7B. NousResearch/Hermes-2-Pro-Mistral-7B matches part of mattshumer/Hermes-2-Pro-11B (291 of its 453 tensors; the rest of mattshumer/Hermes-2-Pro-11B has no counterpart). On the shared part: Partial change: 81% of the weights changed (attention, mlp, norms), in 27 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 54/65, MLP 81/96, Attention 108/128, lm_head 0/1
1 derivatives checked
- MaziyarPanahi/Hermes-2-Pro-11B-GGUF: Quantization