grimjim-merge-workshop/demi-Codex-24B-Small-3.2

Full fine-tune vs Gryphe/Codex-24B-Small-3.2. Fine-tune: 100% of the weights changed, across 40 of 40 layers.

Tensors changed: Embeddings 1/1, Norms 6/81, MLP 120/120, Attention 160/160, lm_head 1/1

1 derivatives checked