whynlp/gsm8k-aug download history

whynlp/gsm8k-aug is a text generation dataset on the Hugging Face Hub. In the last 30 days it was downloaded 1,011 times (529 in the last 7 days), and 29,833 times in total. It ranks #19,149 among datasets by monthly downloads.

GSM8K-AUG This dataset is an augmented version of the GSM8K dataset. It extends the original GSM8K training set to 385k samples by prompting GPT-4. The dataset was originally proposed in paper "From Explicit CoT to Implicit CoT: Learning to Internalize CoT Step by Step". This specific ver

Models trained on gsm8k-aug

14 models list it as training data.

• Lapisbird/Llama-adaLR-model-latent-6-by-1 263 downloads in 30 days
• whynlp/pccot-gpt2 42 downloads in 30 days
• whynlp/pccot-llama1b 40 downloads in 30 days
• Lapisbird/Llama-adaLR-model-latent-6_rl 31 downloads in 30 days
• Lapisbird/Llama-adaLR-model-latent-6-by-1_rl 30 downloads in 30 days
• Lapisbird/Llama-adaLR-model-no_cot_sft 29 downloads in 30 days
• Lapisbird/Llama-adaLR-model-latent-6 28 downloads in 30 days
• Lapisbird/Llama-adaLR-model-cot_sft 21 downloads in 30 days
• Lapisbird/Llama-adaLR-appendix-model-codi_intermediate 19 downloads in 30 days
• Lapisbird/Llama-adaLR-appendix-model-meaned_intermediate 16 downloads in 30 days
• Lapisbird/Llama-adaLR-appendix-model-meaned 16 downloads in 30 days
• Lapisbird/Llama-adaLR-appendix-model-meaned_codi 15 downloads in 30 days

Open whynlp/gsm8k-aug on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.