allenai/tulu-2-dpo-70b
Full fine-tune vs meta-llama/Llama-2-70b-hf. Fine-tune: 100% of the weights changed, across 80 of 80 layers.
Tensors changed: Embeddings 1/1, Norms 161/161, MLP 240/240, Attention 320/320, lm_head 1/1
6 derivatives checked
- TheBloke/tulu-2-dpo-70B-GGUF: Quantization
- mradermacher/tulu-2-dpo-70b-i1-GGUF: Quantization
- mradermacher/tulu-2-dpo-70b-GGUF: Quantization
- nitky/Superswallow-70b-v0.3: Full fine-tune
- Junrulu/FIPO-IPL-IPO-Tulu2-70B: Full fine-tune
- nitky/Superswallow-70b-NVE: Full fine-tune