rl-research/DR-Tulu-SFT-8B
Full fine-tune vs Qwen/Qwen3-8B. Fine-tune: 100% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 1/1, Norms 135/145, MLP 108/108, Attention 144/144, lm_head 1/1
6 derivatives checked
- rl-research/DR-Tulu-8B: Full fine-tune
- Mungert/DR-Tulu-8B-GGUF: Full fine-tune
- mradermacher/DR-Tulu-SFT-8B-i1-GGUF: Quantization
- mradermacher/DR-Tulu-SFT-8B-GGUF: Same weights, converted
- rl-research/DR-Tulu-8B-Step-1900: Full fine-tune
- rl-research/DR-Tulu-No-RLER-8B: Partial fine-tune
1 other repos hold byte-for-byte identical weight files.