cisimcik/cisimcik-base
Full fine-tune (fewer layers) vs Qwen/Qwen3.5-4B-Base. cisimcik/cisimcik-base matches part of Qwen/Qwen3.5-4B-Base (426 of its 738 tensors; the rest of Qwen/Qwen3.5-4B-Base has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, MLP 96/96, Attention 221/224, Norms 105/105
3 derivatives checked
- cisimcik/cisimcik-4b: Full fine-tune
- mradermacher/cisimcik-base-GGUF: Same weights, converted
- cisimcik/tau-4b: Partial fine-tune