legmlai/legml-v1.0-8b
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 121/145, MLP 108/108, Attention 144/144, lm_head 1/1
3 derivatives checked
- mradermacher/legml-v1.0-base-i1-GGUF: Quantization
- mradermacher/legml-v1.0-base-GGUF: Same weights, converted
- legmlai/legml-v1.0-8b-instruct: Full fine-tune