Kiruthikarthi/mistralake-slerp

Full fine-tune (fewer layers) vs HuggingFaceH4/mistral-7b-grok. Kiruthikarthi/mistralake-slerp matches part of HuggingFaceH4/mistral-7b-grok (93 of its 291 tensors; the rest of HuggingFaceH4/mistral-7b-grok has no counterpart). On the shared part: Fine-tune: 91% of the weights changed, across 10 of 10 layers.

Tensors changed: Embeddings 1/1, Norms 13/21, MLP 27/30, Attention 36/40, lm_head 1/1

On the Hub · Downloads on Model Pulse