Kukedlc/NeuralTopBench-7B-ties

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

2 derivatives checked

• mradermacher/NeuralTopBench-7B-ties-GGUF: Same weights, converted
• mradermacher/NeuralTopBench-7B-ties-i1-GGUF: Quantization

On the Hub · Downloads on Model Pulse