CultriX/NeuralTrix-7B-dpo
Full fine-tune vs AiMavenAi/AiMaven-Prometheus. Fine-tune: 100% of the weights changed, across 32 of 32 layers.
Tensors changed: Embeddings 1/1, Norms 65/65, MLP 96/96, Attention 128/128, lm_head 1/1
17 derivatives checked
- mradermacher/NeuralTrix-7B-dpo-i1-GGUF: Quantization
- rizla/rizla-9: Full fine-tune (fewer layers)
- mayacinka/djinn: Full fine-tune
- mradermacher/NeuralTrix-7B-dpo-GGUF: Same weights, converted
- featherless-ai-quants/CultriX-NeuralTrix-7B-dpo-GGUF: Quantization
- tensorblock/CultriX_NeuralTrix-7B-dpo-GGUF: Quantization
- Kukedlc/NeuTrixOmniBe-DPO: Full fine-tune
- Kukedlc/NeuTrixOmniBe-7B-model-remix: Full fine-tune
- CultriX/NeuralTrix-bf16: Full fine-tune
- u66u/NeuralJaskier-7b-dpo: Full fine-tune
- louisbrulenaudet/Pearl-7B-0210-ties: Full fine-tune
- louisbrulenaudet/Pearl-7B-0211-ties: Full fine-tune
- seyf1elislam/KuTrix-7b: Full fine-tune
- jeiku/Paranoid_Android_7B: Full fine-tune
- JDWebProgrammer/Mistral-CultriX-slerp: Full fine-tune
- RaduGabriel/Mistral-7B-Instruct-v0.2NEURAL: Full fine-tune
- mayacinka/NeuralZephyr-Beagle-7B: Full fine-tune