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
1,584
Safetensors
Model size
7.29B params
Tensor type
BF16
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for ezelikman/quietstar-8-ahead

Merges
2 models
Quantizations
1 model

Dataset used to train ezelikman/quietstar-8-ahead