grimjim/Llama-3.1-SuperNova-Lite-lorabilterated-8B
Partial fine-tune vs arcee-ai/Llama-3.1-SuperNova-Lite. Partial change: 31% of the weights changed (attention, mlp, norms), in 32 of 32 layers.
Tensors changed: Embeddings 0/1, Norms 3/65, MLP 32/96, Attention 46/128, lm_head 0/1
11 derivatives checked
- mradermacher/Llama-3.1-SuperNova-Lite-lorabilterated-8B-i1-GGUF: Quantization
- featherless-ai-quants/grimjim-Llama-3.1-SuperNova-Lite-lorabilterated-8B-GGUF: Quantization
- mradermacher/Llama-3.1-SuperNova-Lite-lorabilterated-8B-GGUF: Quantization
- QuantFactory/NovaSpark-GGUF: Full fine-tune
- Epiculous/NovaSpark-GGUF: Full fine-tune
- Epiculous/NovaSpark: Full fine-tune
- BoltMonkey/NeuralDaredevil-SuperNova-Lite-7B-DARETIES-abliterated: Full fine-tune
- Nexesenex/Llama_3.1_8b_Smarteaz_0.21_SN: Full fine-tune
- Nexesenex/Llama_3.1_8b_Smarteaz_0.11a: Full fine-tune
- Nexesenex/Llama_3.x_8b_Smarteaz_0.1a: Partial fine-tune
- Nexesenex/Llama_3.1_8b_Smarteaz_0.21_R1: Full fine-tune