morganstanley/qqWen-14B-RL-Reasoning
Full fine-tune vs Qwen/Qwen2.5-14B-Instruct. Fine-tune: 95% of the weights changed, across 48 of 48 layers.
Tensors changed: Embeddings 0/1, Norms 94/97, MLP 144/144, Attention 336/336, lm_head 1/1
3 derivatives checked
- mradermacher/qqWen-14B-RL-Reasoning-i1-GGUF: Quantization
- mradermacher/qqWen-14B-RL-Reasoning-GGUF: Quantization
- LogicBombaklot/qqWen-14B-RL-Reasoning-mlx-8Bit: Quantization
1 other repos hold byte-for-byte identical weight files.