arcee-ai/Arcee-VyLinh
Partial fine-tune vs Qwen/Qwen2.5-3B-Instruct. Partial change: 65% of the weights changed (attention, embeddings, mlp, norms), in 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 60/73, MLP 68/108, Attention 155/252
4 derivatives checked
- mradermacher/Arcee-VyLinh-i1-GGUF: Quantization
- mradermacher/Arcee-VyLinh-GGUF: Same weights, converted
- MaziyarPanahi/Arcee-VyLinh-GGUF: Same weights, converted
- DevQuasar-3/arcee-ai.Arcee-VyLinh-GGUF: Quantization