google/gemma-2-27b-it
Full fine-tune vs google/gemma-2-27b. Fine-tune: 100% of the weights changed, across 46 of 46 layers.
Tensors changed: Embeddings 1/1, Attention 184/184, Norms 185/185, MLP 138/138
31 derivatives checked
- bartowski/gemma-2-27b-it-GGUF: Quantization
- legraphista/gemma-2-27b-it-IMat-GGUF: Same weights, converted
- Skywork/Skywork-Reward-Gemma-2-27B-v0.2: Full fine-tune
- lmstudio-community/gemma-2-27b-it-GGUF: Quantization
- mradermacher/gemma-2-27b-it-GGUF: Quantization
- yujiepan/gemma-2-tiny-random: Different architecture
- mradermacher/gemma-2-27b-it-i1-GGUF: Quantization
- qwp4w3hyb/gemma-2-27b-it-iMat-GGUF: Quantization
- second-state/gemma-2-27b-it-GGUF: Same weights, converted
- Saxo/Linkbricks-Horizon-AI-Korean-Gemma-2-sft-dpo-27B: LoRA-style
- mbley/google-gemma-2-27b-it-AWQ: Quantization
- MarsupialAI/Big-Tiger-Gemma-27B-v1_iMatrix_GGUF: LoRA-style
- Saxo/Linkbricks-Horizon-AI-Korean-Advanced-27B: LoRA-style
- shenzhi-wang/Gemma-2-27B-Chinese-Chat: Full fine-tune
- gaianet/gemma-2-27b-it-GGUF: Same weights, converted
- Skywork/Skywork-Reward-Gemma-2-27B: Full fine-tune
- DiTy/gemma-2-27b-it-function-calling-GGUF: LoRA-style
- afg1/gemma-2-27b-it: Same weights, converted
- google/datagemma-rig-27b-it: Full fine-tune
- google/datagemma-rag-27b-it: Full fine-tune
- byroneverson/gemma-2-27b-it-abliterated: Abliterated
- AALF/gemma-2-27b-it-SimPO-37K: Full fine-tune
- Alfaxad/gemma2-27b-swahili-it: LoRA-style
- AALF/gemma-2-27b-it-SimPO-37K-100steps: Full fine-tune
- MBZUAI-Paris/Atlas-Chat-27B: LoRA-style
- nbeerbower/gemma2-gutenberg-27B: LoRA-style
- Saxo/Linkbricks-Horizon-AI-Korean-Superb-27B: Full fine-tune
- jwang2373/UW-SBEL-ChronoGemma-27b-it: LoRA-style
- prithivMLmods/Blaze.1-27B-Preview: LoRA-style
- prithivMLmods/Blaze.1-27B-Reflection: LoRA-style
10 other repos hold byte-for-byte identical weight files.