puwaer/DeepSeek-V4-Flash-0731-reap-200b
Partial fine-tune (fewer layers) vs deepseek-ai/DeepSeek-V4-Flash-0731. puwaer/DeepSeek-V4-Flash-0731-reap-200b matches part of deepseek-ai/DeepSeek-V4-Flash-0731 (47405 of its 72317 tensors; the rest of deepseek-ai/DeepSeek-V4-Flash-0731 has no counterpart). On the shared part: Partial change: 19% of the weights changed (mlp), in 43 of 43 layers.
Tensors changed: Embeddings 0/1, MLP 877/1274, Attention 0/851, Norms 0/235, other 0/4
1 derivatives checked
- puwaer/DeepSeek-V4-Flash-0731-reap-200b-gguf: Not comparable yet