chihoonlee10/T3Q-Mistral-Orca-Math-DPO
LoRA-style vs liminerity/M7-7b. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, up_proj changed, in 32 of 32 layers; norms, biases and embeddings are untouched.
Tensors changed: MLP 96/96, Attention 0/128, Norms 0/65, Embeddings 0/1, lm_head 0/1
10 derivatives checked
- mradermacher/T3Q-Mistral-Orca-Math-DPO-GGUF: Quantization
- nbeerbower/bruphin-lambda: Full fine-tune
- automerger/Experiment28T3q-7B: Full fine-tune
- automerger/T3qM7-7B: LoRA-style
- automerger/Experiment26T3q-7B: Full fine-tune
- automerger/ShadowT3q-7B: Full fine-tune
- automerger/T3qm7xpT3q-7B: Full fine-tune
- automerger/Strangemerges_32T3q-7B: Full fine-tune
- automerger/YamshadowT3q-7B: Full fine-tune
- automerger/CalmexperimentT3q-7B: Full fine-tune