danieldk/Qwen2.5-1.5B-Instruct-w8a8-int-dynamic-weight

Full fine-tune vs Qwen/Qwen2.5-1.5B. Fine-tune: 100% of the weights changed, across 28 of 28 layers.

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

1 derivatives checked

• featherless-ai-quants/danieldk-Qwen2.5-1.5B-Instruct-w8a8-int-dynamic-weight-GGUF: Quantization

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

On the Hub · Downloads on Model Pulse