steinzer-narayan/fimbulhermes-15B-v0.1
Partial fine-tune (fewer layers) vs NousResearch/Nous-Hermes-2-SOLAR-10.7B. NousResearch/Nous-Hermes-2-SOLAR-10.7B matches part of steinzer-narayan/fimbulhermes-15B-v0.1 (435 of its 624 tensors; the rest of steinzer-narayan/fimbulhermes-15B-v0.1 has no counterpart). On the shared part: Partial change: 65% of the weights changed (attention, mlp, norms), in 32 of 48 layers.
Tensors changed: Embeddings 0/1, Norms 64/97, MLP 96/144, Attention 128/192, lm_head 0/1
1 derivatives checked
- mradermacher/fimbulhermes-15B-v0.1-GGUF: Quantization