cerebras/DeepSeek-V3.2-REAP-508B-A37B
Partial fine-tune vs deepseek-ai/DeepSeek-V3.2. Partial change: 25% of the weights changed (mlp), in 58 of 61 layers.
Tensors changed: Embeddings 0/1, Attention 0/915, Norms 0/367, MLP 883/1260, lm_head 0/1
Partial fine-tune vs deepseek-ai/DeepSeek-V3.2. Partial change: 25% of the weights changed (mlp), in 58 of 61 layers.
Tensors changed: Embeddings 0/1, Attention 0/915, Norms 0/367, MLP 883/1260, lm_head 0/1