Kukedlc/Neural-4-Maths-7b

Full fine-tune vs Kukedlc/Neural4gsm8k. Fine-tune: 100% of the weights changed, across 32 of 32 layers.

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

3 derivatives checked

• mradermacher/Neural-4-Maths-7b-i1-GGUF: Quantization
• featherless-ai-quants/Kukedlc-Neural-4-Maths-7b-GGUF: Quantization
• mradermacher/Neural-4-Maths-7b-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse