mayacinka/yam-jom-7B-slerp
Full fine-tune vs eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO-v2. Fine-tune: 97% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 65/65, MLP 93/96, Attention 124/128, lm_head 1/1
4 derivatives checked
- mradermacher/yam-jom-7B-slerp-GGUF: Same weights, converted
- automerger/Experiment28Yam-7B: Full fine-tune
- automerger/YamExperiment24-7B: Full fine-tune
- automerger/Experiment27Yam-7B: Full fine-tune