newmindai/Mursit-Embed-Qwen3-1.7B-TR
Partial fine-tune vs newmindai/Mecellem-Qwen3-1.7B-TR. Partial change: 82% of the weights changed (attention, mlp, norms), in 28 of 28 layers.
Tensors changed: Embeddings 0/1, Norms 84/113, MLP 84/84, Attention 112/112
2 derivatives checked
- mradermacher/Mursit-Embed-Qwen3-1.7B-TR-i1-GGUF: Quantization
- mradermacher/Mursit-Embed-Qwen3-1.7B-TR-GGUF: Same weights, converted