SteelStorage/VerB-Etheria-55b
Partial fine-tune (fewer layers) vs brucethemoose/Yi-34B-200K-DARE-megamerge-v8. brucethemoose/Yi-34B-200K-DARE-megamerge-v8 matches part of SteelStorage/VerB-Etheria-55b (543 of its 885 tensors; the rest of SteelStorage/VerB-Etheria-55b has no counterpart). On the shared part: Partial change: 75% of the weights changed (attention, mlp, norms), in 46 of 60 layers.
Tensors changed: Embeddings 0/1, Norms 92/121, MLP 138/180, Attention 184/240, lm_head 0/1
1 derivatives checked
- mradermacher/VerB-Etheria-55b-GGUF: Quantization