Myungkyu/rldx_1_rmbench_taco_wodemo_gemini_b128_60k

Partial fine-tune vs RLWRLD/RLDX-1-PT. Partial change: 29% of the weights changed (attention, embeddings, mlp, norms, other), in 4 of 27 layers.

Tensors changed: other 56/69, Embeddings 4/9, Norms 48/221, MLP 60/226, Attention 32/196, lm_head 0/1

On the Hub · Downloads on Model Pulse