HoangHa/Pensez-v0.1-e5
Full fine-tune vs Qwen/Qwen2.5-7B-Instruct. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Norms 52/57, MLP 84/84, Attention 196/196
4 derivatives checked
- mradermacher/Pensez-v0.1-e5-i1-GGUF: Quantization
- DevQuasar-9/HoangHa.Pensez-v0.1-e5-GGUF: Same weights, converted
- mradermacher/Pensez-v0.1-e5-GGUF: Same weights, converted
- Xiaojian9992024/Qwen2.5-7B-MS-Destroyer: Full fine-tune