tetracta/Chimera-1-Qwen2.5-3B

Partial fine-tune vs Qwen/Qwen2.5-3B-Instruct. Partial change: 20% of the weights changed (attention, mlp, norms), in 8 of 36 layers.

Tensors changed: Embeddings 0/1, Norms 9/73, MLP 24/108, Attention 54/252

1 derivatives checked

• mradermacher/Chimera-1-Qwen2.5-3B-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse