usermma/LFM2.5-230M-ECE-OBLITERATED
Partial fine-tune vs usermma/LFM2.5-230M-ECE. Partial change: 31% of the weights changed (attention, embeddings), in 1 of 14 layers.
Tensors changed: Embeddings 1/2, other 0/24, MLP 0/42, Norms 0/40, Attention 4/24
2 derivatives checked
- mradermacher/LFM2.5-230M-ECE-OBLITERATED-i1-GGUF: Quantization
- mradermacher/LFM2.5-230M-ECE-OBLITERATED-GGUF: Same weights, converted