qingy2024/Ling-Mini-2.0-Identity

Partial fine-tune vs inclusionAI/Ling-mini-2.0. Partial change: 82% of the weights changed (attention, lm_head, mlp, norms), in 20 of 20 layers.

Tensors changed: MLP 2498/2877, Embeddings 0/1, Attention 40/40, lm_head 1/1, Norms 25/81

1 derivatives checked

• mradermacher/Ling-Mini-2.0-Identity-GGUF: Quantization

On the Hub · Downloads on Model Pulse