jerrimu/ERNIE-21B-REAP

Full fine-tune vs baidu/ERNIE-4.5-21B-A3B-Base-PT. Fine-tune: 100% of the weights changed, across 28 of 28 layers.

Tensors changed: Embeddings 1/1, Norms 57/57, MLP 2269/2269, Attention 112/112

1 derivatives checked

• mradermacher/ERNIE-21B-REAP-GGUF: Quantization

On the Hub · Downloads on Model Pulse