google/gemma-2-2b-jpn-it
Full fine-tune vs google/gemma-2-2b-it. Fine-tune: 100% of the weights changed, across 26 of 26 layers.
Tensors changed: Embeddings 1/1, Norms 105/105, MLP 78/78, Attention 104/104
21 derivatives checked
- mradermacher/gemma-2-2b-jpn-it-i1-GGUF: Quantization
- bartowski/gemma-2-2b-jpn-it-GGUF: Same weights, converted
- QuantFactory/EZO-gemma-2-2b-jpn-it-GGUF: Quantization
- dahara1/gemma-2-2b-jpn-it-gguf-japanese-imatrix: Quantization
- webbigdata/gemma-2-2b-jpn-it-translate-gguf: Quantization
- MCZK/gemma-2-2b-jpn-it-GGUF: Same weights, converted
- mradermacher/gemma-2-2b-jpn-it-GGUF: Same weights, converted
- ymcki/gemma-2-2b-jpn-it-GGUF: Same weights, converted
- alfredplpl/gemma-2-2b-jpn-it-gguf: Same weights, converted
- DevQuasar-5/google.gemma-2-2b-jpn-it-GGUF: Quantization
- kun432/Google-gemma-2-2b-jpn-it-mlx-4bit: Quantization
- webbigdata/gemma-2-2b-jpn-it-translate: LoRA-style
- ymcki/gemma-2-2b-jpn-it-abliterated-18: Abliterated
- ymcki/gemma-2-2b-ORPO-jpn-it-abliterated-18: LoRA-style
- ymcki/gemma-2-2b-jpn-it-abliterated-17: Abliterated
- ymcki/gemma-2-2b-jpn-it-abliterated-17-ORPO: LoRA-style
- ymcki/gemma-2-2b-jpn-it-abliterated-24: Abliterated
- ymcki/gemma-2-2b-jpn-it-abliterated-18-ORPO: LoRA-style
- AXCXEPT/EZO-gemma-2-2b-jpn-it: Full fine-tune
- Sakalti/SJT-2B: Same weights, converted
- fyuuki0jp/gemma-2-2b-jpn-it_finetuning_sft: Partial fine-tune