sillykiwi/Qwen3-4B-PastelCoder

Full fine-tune vs prithivMLmods/Blitzar-Coder-4B-F.1. Fine-tune: 100% of the weights changed, across 36 of 36 layers.

Tensors changed: Norms 81/145, MLP 108/108, Attention 144/144

2 derivatives checked

• mradermacher/Qwen3-4B-PastelCoder-i1-GGUF: Quantization
• mradermacher/Qwen3-4B-PastelCoder-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse