dphn/Dolphin3.0-R1-Mistral-24B
Full fine-tune vs mistralai/Mistral-Small-24B-Base-2501. Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: Norms 81/81, MLP 120/120, Attention 160/160
12 derivatives checked
- bartowski/cognitivecomputations_Dolphin3.0-R1-Mistral-24B-GGUF: Quantization
- eaddario/Dolphin3.0-R1-Mistral-24B-GGUF: Same weights, converted
- mradermacher/Dolphin3.0-R1-Mistral-24B-i1-GGUF: Quantization
- mradermacher/Dolphin3.0-R1-Mistral-24B-GGUF: Quantization
- mlx-community/Dolphin3.0-R1-Mistral-24B-4bit: Quantization
- SAIFIINDUSTRIES/Abliterated-Dolphin3.0-R1-Mistral-24B: Abliterated
- DevQuasar-7/cognitivecomputations.Dolphin3.0-R1-Mistral-24B-GGUF: Same weights, converted
- Valdemardi/Dolphin3.0-R1-Mistral-24B-AWQ: Quantization
- tensorblock/Dolphin3.0-R1-Mistral-24B-GGUF: Quantization
- pki/nova-24b-cybersec: LoRA-style
- saukko/Abliterated-Dolphin3.0-R1-Mistral-24B: Abliterated
- AlSamCur123/DolphinMistral-24BContinuedFine: LoRA-style
1 other repos hold byte-for-byte identical weight files.