perplexity-ai/pplx-decider-v1-27b

Full fine-tune vs Qwen/Qwen3.8-27B. Fine-tune: 95% of the weights changed, across 64 of 64 layers.

Tensors changed: Norms 180/319, Attention 465/556, MLP 272/304, Embeddings 2/4

1 derivatives checked

• ramgpt/pplx-decider-v1-27b-EXL3: Same weights, converted

On the Hub · Downloads on Model Pulse