automerger/Experiment26T3q-7B

Full fine-tune vs chihoonlee10/T3Q-Mistral-Orca-Math-DPO. Fine-tune: 97% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 1/1, Norms 18/65, MLP 93/96, Attention 124/128, lm_head 1/1

2 derivatives checked

• MaziyarPanahi/Experiment26T3q-7B-GGUF: Quantization
• MaziyarPanahi/MeliodasPercival_01_Experiment26T3q: Full fine-tune

On the Hub · Downloads on Model Pulse