nllg/EdiTikZ-4B-RL

Partial fine-tune vs nllg/EdiTikZ-4B. Partial change: 79% of the weights changed (attention, mlp, norms), in 32 of 32 layers.

Tensors changed: Embeddings 0/4, Norms 68/203, Attention 186/320, MLP 96/196

1 derivatives checked

• mradermacher/EdiTikZ-4B-RL-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse