OpenGVLab/InternVL3-1B
Full fine-tune vs OpenGVLab/InternVL3-1B-Instruct. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: lm_head 1/1, Embeddings 5/5, Norms 65/145, MLP 170/174, Attention 249/264, other 0/48
3 derivatives checked
- unsloth/InternVL3-1B-GGUF: Same weights, converted (fewer layers)
- mradermacher/InternVL3-1B-GGUF: Same weights, converted (fewer layers)
- mradermacher/InternVL3-1B-i1-GGUF: Quantization (fewer layers)
3 other repos hold byte-for-byte identical weight files.