allenai/tmax-8b

Full fine-tune vs allenai/tmax-sft-8b. Fine-tune: 92% of the weights changed, across 36 of 36 layers.

Tensors changed: lm_head 1/1, Embeddings 0/1, Norms 97/145, MLP 108/108, Attention 144/144

2 derivatives checked

• mradermacher/tmax-8b-i1-GGUF: Quantization
• mradermacher/tmax-8b-GGUF: Same weights, converted

On the Hub · Downloads on Model Pulse