KatyTheCutie/LemonadeRP-4.5.3
Full fine-tune vs NeverSleep/Noromaid-7B-0.4-DPO. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Norms 65/65, MLP 96/96, Attention 128/128
18 derivatives checked
- mradermacher/LemonadeRP-4.5.3-i1-GGUF: Quantization
- backyardai/LemonadeRP-4.5.3-GGUF: Same weights, converted
- mradermacher/LemonadeRP-4.5.3-GGUF: Same weights, converted
- QuantFactory/Chewy-Lemon-Cookie-11B-GGUF: Full fine-tune (fewer layers)
- grimjim/kukulemon-7B: Full fine-tune
- grimjim/kunoichi-lemon-royale-7B: Full fine-tune
- MisterRaven006/SweetNeural-7B: Full fine-tune
- FallenMerick/Smart-Lemon-Cookie-7B: Full fine-tune
- FallenMerick/Chewy-Lemon-Cookie-11B: Full fine-tune (fewer layers)
- FallenMerick/Chunky-Lemon-Cookie-11B: Full fine-tune (fewer layers)
- head-empty-ai/Mytho-Lemon-11B: Partial fine-tune (fewer layers)
- FallenMerick/Big-Lemon-Cookie-11B: Full fine-tune (fewer layers)
- ABX-AI/Kuno-Lemon-7B: Full fine-tune
- FallenMerick/Iced-Lemon-Cookie-7B: Full fine-tune
- mpasila/LemonadeRP-4.5.3-11B: Partial fine-tune (fewer layers)
- itchindigo/NeuralLemon: Full fine-tune
- ABX-AI/Laymonade-7B: Full fine-tune
- Statuo/LemonWizardv3: Full fine-tune
2 other repos hold byte-for-byte identical weight files.