sethuiyer/Chikuma_10.7B
Full fine-tune (fewer layers) vs openchat/openchat-3.5-0106. openchat/openchat-3.5-0106 matches part of sethuiyer/Chikuma_10.7B (291 of its 435 tensors; the rest of sethuiyer/Chikuma_10.7B has no counterpart). On the shared part: Fine-tune: 98% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 64/65, MLP 96/96, Attention 128/128, lm_head 0/1
3 derivatives checked
- mradermacher/Chikuma_10.7B-i1-GGUF: Quantization
- mradermacher/Chikuma_10.7B-GGUF: Same weights, converted
- sethuiyer/Chikuma_10.7B_v2: LoRA-style