google/gemma-3-270m-it-qat-q4_0-unquantized
Full fine-tune vs google/gemma-3-270m-it. Fine-tune: 100% of the weights changed, across 18 of 18 layers.
Tensors changed: Embeddings 1/1, Norms 106/109, MLP 54/54, Attention 72/72
5 derivatives checked
- unsloth/gemma-3-270m-it-qat-GGUF: Same weights, converted
- bartowski/google_gemma-3-270m-it-qat-GGUF: Same weights, converted
- lmstudio-community/gemma-3-270m-it-qat-MLX-4bit: Quantization
- lmstudio-community/gemma-3-270m-it-qat-GGUF: Quantization
- unsloth/gemma-3-270m-it-qat: Identical copy
1 other repos hold byte-for-byte identical weight files.