Sakalti/SJT-24B-Alpha
Partial fine-tune (fewer layers) vs Sakalti/ultiima-14B-v0.4. Sakalti/ultiima-14B-v0.4 matches part of Sakalti/SJT-24B-Alpha (579 of its 987 tensors; the rest of Sakalti/SJT-24B-Alpha has no counterpart). On the shared part: Partial change: 67% of the weights changed (attention, mlp, norms), in 36 of 48 layers.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 72/97, MLP 108/144, Attention 252/336
1 derivatives checked
- mradermacher/SJT-24B-Alpha-i1-GGUF: Quantization