bertin-project/Gromenauer-7B-Instruct

Full fine-tune vs bertin-project/Gromenauer-7B. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

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

1 derivatives checked

• mradermacher/Gromenauer-7B-Instruct-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse