Datasets:
Formats:
json
Size:
10K - 100K
Tags:
creative-writing
narrative-reasoning
long-form-fiction
chain-of-thought
multilingual
synthetic-data
DOI:
License:
Model confir
#2
by Mikoris - opened
Hello, I notice that this dataset was generate by a 1T model. Could you tell me what the model is, so i can verify the quality of this database, then apply it to model finetune. Thank you very much!
Hello! This dataset was distilled from MiMo v2.5 when it was a stealth model. I don't trust the quality that much because ai models are terrible at loading a meaning to word. You can still use this dataset for your model. I want to see how it will handle writing. When you finish training, can you perform a benchmark or something? I really want to see how this dataset will impact.
Have a good day and training!
-Dxniz