M4-ai/TinyMistral-6x248M-Instruct

Full fine-tune vs M4-ai/TinyMistral-6x248M. Fine-tune: 100% of the weights changed, across 12 of 12 layers.

Tensors changed: lm_head 1/1, Embeddings 1/1, MLP 216/216, other 12/12, Norms 25/25, Attention 48/48

1 derivatives checked

• mradermacher/TinyMistral-6x248M-Instruct-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse