ByteDance-Seed/cudaLLM-8B
Full fine-tune vs Qwen/Qwen3-8B. Fine-tune: 92% of the weights changed, across 36 of 36 layers.
Tensors changed: Embeddings 0/1, Norms 128/145, MLP 108/108, Attention 144/144, lm_head 1/1
4 derivatives checked
- mradermacher/cudaLLM-8B-i1-GGUF: Quantization
- prithivMLmods/cudaLLM-8B-GGUF: Partial fine-tune
- mradermacher/cudaLLM-8B-GGUF: Same weights, converted
- DevQuasar/ByteDance-Seed.cudaLLM-8B-GGUF: Same weights, converted