abeja/ABEJA-Qwen2.5-7b-Japanese-v0.1
Full fine-tune vs Qwen/Qwen2.5-7B-Instruct. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 57/57, MLP 84/84, Attention 196/196, lm_head 1/1
5 derivatives checked
- tatsuyaaaaaaa/ABEJA-Qwen2.5-7b-Japanese-v0.1-gguf: Same weights, converted
- mmnga/ABEJA-Qwen2.5-7b-Japanese-v0.1-gguf: Quantization
- mradermacher/ABEJA-Qwen2.5-7b-Japanese-v0.1-i1-GGUF: Quantization
- mradermacher/ABEJA-Qwen2.5-7b-Japanese-v0.1-GGUF: Same weights, converted
- featherless-ai-quants/abeja-ABEJA-Qwen2.5-7b-Japanese-v0.1-GGUF: Quantization