mergekit-community/L3.1-Artemis-dcd-12B
Full fine-tune (fewer layers) vs mergekit-community/L3.1-Artemis-c-8B. mergekit-community/L3.1-Artemis-c-8B matches part of mergekit-community/L3.1-Artemis-dcd-12B (291 of its 435 tensors; the rest of mergekit-community/L3.1-Artemis-dcd-12B has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 65/65, MLP 96/96, Attention 128/128
2 derivatives checked
- mradermacher/L3.1-Artemis-dcd-12B-i1-GGUF: Quantization
- mradermacher/L3.1-Artemis-dcd-12B-GGUF: Quantization