nbeerbower/EVA-Rombos1-Qwen2.5-32B-lorablated
Partial fine-tune vs nbeerbower/EVA-Rombos1-Qwen2.5-32B. Partial change: 41% of the weights changed (attention, embeddings, mlp, norms), in 64 of 64 layers.
Tensors changed: lm_head 0/1, Embeddings 1/1, Norms 128/129, MLP 77/192, Attention 87/448
2 derivatives checked
- mradermacher/EVA-Rombos1-Qwen2.5-32B-lorablated-i1-GGUF: Quantization
- mradermacher/EVA-Rombos1-Qwen2.5-32B-lorablated-GGUF: Quantization