| datasets: | |
| - open-web-math/open-web-math | |
| Mistral-7b with continued pretraining using Quiet-STaR (https://arxiv.org/abs/2403.09629) for generating 8 thought tokens before each output token. | |
| datasets: | |
| - open-web-math/open-web-math | |
| Mistral-7b with continued pretraining using Quiet-STaR (https://arxiv.org/abs/2403.09629) for generating 8 thought tokens before each output token. | |