Kukedlc/NeuralKrishna-7B-v3
Full fine-tune vs Kukedlc/Fasciculus-Arcuatus-7B-slerp. 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
4 derivatives checked
- mradermacher/NeuralKrishna-7B-v3-i1-GGUF: Quantization
- bartowski/NeuralSirKrishna-7b-GGUF: Full fine-tune
- mradermacher/NeuralKrishna-7B-v3-GGUF: Same weights, converted
- Kukedlc/NeuralSirKrishna-7b: Full fine-tune