Mistral-7b with continued pretraining using Quiet-STaR (https://arxiv.org/abs/2403.09629) for generating 8 thought tokens before each output token.
- Downloads last month
- 9
Mistral-7b with continued pretraining using Quiet-STaR (https://arxiv.org/abs/2403.09629) for generating 8 thought tokens before each output token.