google/gemma-7b-it
Full fine-tune vs google/gemma-7b. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 57/57, MLP 84/84, Attention 112/112
12 derivatives checked
- MaziyarPanahi/gemma-7b-it-GGUF: Same weights, converted
- mradermacher/gemma-7b-it-GGUF: Same weights, converted
- mradermacher/gemma-7b-it-i1-GGUF: Quantization
- second-state/Gemma-7b-it-GGUF: Quantization
- sayhan/gemma-7b-it-GGUF-quantized: Same weights, converted
- brittlewis12/gemma-7b-it-GGUF: Same weights, converted
- QueryloopAI/gemma-7b-openhermes: LoRA-style
- lemon-mint/gemma-ko-7b-it-v0.40: LoRA-style
- abideen/gemma-7b-openhermes: LoRA-style
- yhkim9362/gemma-en-ko-7b-v0.2: Identical copy
- yhkim9362/gemma-en-ko-7b-v0.1: Identical copy
- ihopper/ko-gemma-7b-sft-dpo-v1.0: LoRA-style
17 other repos hold byte-for-byte identical weight files.