Dataset description
This release incorporates the complete data sequence used in our CrystalCoder training, covering data sequences from the three pre-training stages. The data is a combination of two previous works: SlimPajama dataset and StarCoder. We divide them across three stages with different weights.
Stage 1
During this stage, we utilize half of the SlimPajama data.
Stage 2
In the second stage, the remaining half of the SlimPajama data is employed, along with two epochs of StarCoder data. For StarCoder data, we implement FIM augmentation with an FIM rate of 0.9.
Stage 3
The third stage involves the reuse of Python, HTML, CSS, and JavaScript data from the StarCoder data for training over three epochs. In this stage, FIM with a rate of 0.3 is applied. Additionally, a small portion of the SlimPajama dataset except the Github part is also reused.
Primary usage
This dataset is used for our training of our CrystalCoder, and also for further reproduction.