rl-research/DR-Tulu-8B-Step-1900
Full fine-tune vs rl-research/DR-Tulu-SFT-8B. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 95/145, MLP 108/108, Attention 144/144, lm_head 1/1
2 derivatives checked
- mradermacher/DR-Tulu-8B-Step-1900-i1-GGUF: Quantization
- mradermacher/DR-Tulu-8B-Step-1900-GGUF: Same weights, converted
1 other repos hold byte-for-byte identical weight files.