Kukedlc/NeuralGanesha-7b

Full fine-tune vs Kukedlc/MyModelsMerge-7b. Fine-tune: 99% of the weights changed, across 32 of 32 layers.

Tensors changed: Embeddings 1/1, Norms 65/65, MLP 96/96, Attention 124/128, lm_head 1/1

2 derivatives checked

• mradermacher/NeuralGanesha-7b-GGUF: Quantization
• Kukedlc/Ramakrishna-7b-v4: Full fine-tune

On the Hub · Downloads on Model Pulse