zai-org/webrl-glm-4-9b
Full fine-tune vs zai-org/glm-4-9b. Fine-tune: 93% of the weights changed, across 40 of 40 layers.
Tensors changed: Embeddings 0/1, Norms 23/81, MLP 80/80, Attention 120/120, other 1/2
Full fine-tune vs zai-org/glm-4-9b. Fine-tune: 93% of the weights changed, across 40 of 40 layers.
Tensors changed: Embeddings 0/1, Norms 23/81, MLP 80/80, Attention 120/120, other 1/2