Kukedlc/NeuralKrishna-7B-V2

Full fine-tune vs Kukedlc/Neural4gsm8k. Fine-tune: 97% of the weights changed, across 32 of 32 layers.

Tensors changed: lm_head 1/1, Norms 65/65, MLP 93/96, Attention 124/128, Embeddings 1/1

1 derivatives checked

• mradermacher/NeuralKrishna-7B-V2-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse