win10/GLM-4-T1-54B
Full fine-tune (fewer layers) vs zai-org/GLM-4-32B-0414. zai-org/GLM-4-32B-0414 matches part of win10/GLM-4-T1-54B (613 of its 1043 tensors; the rest of win10/GLM-4-T1-54B has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 61 of 61 layers.
Tensors changed: Embeddings 1/1, Norms 245/245, MLP 122/122, Attention 244/244, lm_head 1/1
2 derivatives checked
- mradermacher/GLM-4-T1-54B-i1-GGUF: Quantization
- mradermacher/GLM-4-T1-54B-GGUF: Quantization