juancopi81/lmd-8bars-2048-epochs40_v4

Full fine-tune vs juancopi81/lmd-8bars-2048-epochs30_v4. Fine-tune: 100% of the weights changed, across 6 of 6 layers.

Tensors changed: Attention 24/24, Norms 26/26, MLP 24/24, Embeddings 2/2

3 other repos hold byte-for-byte identical weight files.

On the Hub · Downloads on Model Pulse