apple/DiffuCoder-7B-cpGRPO

Full fine-tune vs apple/DiffuCoder-7B-Instruct. Fine-tune: 93% of the weights changed, across 28 of 28 layers.

Tensors changed: Embeddings 0/1, Norms 46/57, MLP 84/84, Attention 196/196, lm_head 1/1

2 derivatives checked

• mradermacher/DiffuCoder-7B-cpGRPO-i1-GGUF: Quantization
• mradermacher/DiffuCoder-7B-cpGRPO-GGUF: Same weights, converted

1 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse