Lambent/Apeira-v0-27B

Full fine-tune vs Lambent/Mira-v1.11-Ties-27B. Fine-tune: 100% of the weights changed, across 62 of 62 layers.

Tensors changed: Embeddings 4/4, Norms 483/484, MLP 291/294, Attention 464/464, other 1/1

2 derivatives checked

• mradermacher/Apeira-v0-27B-i1-GGUF: Quantization (fewer layers)
• mradermacher/Apeira-v0-27B-GGUF: Quantization (fewer layers)

On the Hub · Downloads on Model Pulse