EVA-UNIT-01/EVA-Qwen2.5-14B-v0.0
Partial fine-tune vs Qwen/Qwen2.5-14B. Partial change: 50% of the weights changed (attention, lm_head, mlp), in 45 of 48 layers.
Tensors changed: Embeddings 0/1, Norms 0/97, MLP 72/144, Attention 168/336, lm_head 1/1
4 derivatives checked
- bartowski/EVA-Qwen2.5-14B-v0.0-GGUF: Same weights, converted
- mradermacher/EVA-Qwen2.5-14B-v0.0-i1-GGUF: Quantization
- mradermacher/EVA-Qwen2.5-14B-v0.0-GGUF: Quantization
- CultriX/Qwen2.5-14B-Wernicke: Full fine-tune