fblgit/miniclaus-qw1.5B-UNAMGS-GRPO
Partial fine-tune vs fblgit/miniclaus-qw1.5B-UNAMGS. Partial change: 85% of the weights changed (attention, mlp, norms), in 28 of 28 layers.
Tensors changed: Embeddings 0/1, Norms 32/57, MLP 84/84, Attention 182/196
2 derivatives checked
- bartowski/fblgit_miniclaus-qw1.5B-UNAMGS-GRPO-GGUF: Same weights, converted
- mradermacher/miniclaus-qw1.5B-UNAMGS-GRPO-GGUF: Same weights, converted