DavidAU/LFM2-8B-A1B-SpeedDemon-The-Deckard-II-Series-A

Partial fine-tune vs LiquidAI/LFM2-8B-A1B. Partial change: 5% of the weights changed (attention, mlp, other), in 24 of 24 layers.

Tensors changed: MLP 28/2162, Embeddings 0/2, other 36/54, Norms 0/60, Attention 24/24

2 derivatives checked

• mradermacher/LFM2-8B-A1B-SpeedDemon-The-Deckard-II-Series-A-i1-GGUF: Quantization
• mradermacher/LFM2-8B-A1B-SpeedDemon-The-Deckard-II-Series-A-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse