usermma/Atomight-V2.2-UltraThink-0.5B-OBLITERATED
Partial fine-tune vs NovatasticRoScript/Atomight-V2.2-UltraThink-0.5B. Partial change: 40% of the weights changed (attention, embeddings, mlp), in 4 of 24 layers.
Tensors changed: Embeddings 1/1, Norms 0/49, MLP 12/72, Attention 20/168
2 derivatives checked
- mradermacher/Atomight-V2.2-UltraThink-0.5B-OBLITERATED-i1-GGUF: Quantization
- mradermacher/Atomight-V2.2-UltraThink-0.5B-OBLITERATED-GGUF: Same weights, converted