ConvexAI/Metabird-7B

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

Tensors changed: Norms 13/65, MLP 96/96, Attention 128/128

3 derivatives checked

• mradermacher/Metabird-7B-i1-GGUF: Quantization
• mradermacher/Metabird-7B-GGUF: Same weights, converted
• NovoCode/Metabird-7b-DPO: Full fine-tune

On the Hub · Downloads on Model Pulse