Datasets:

distily
/

synth_gpt2_t1_seq_100K

Tags:

Dataset card Files Files and versions

synth_gpt2_t1_seq_100K / README.md

lapp0's picture

Upload README.md with huggingface_hub

4a1719c verified over 1 year ago

|

history blame contribute delete

342 Bytes

source_datasets:
  - Original
  - Synthetic
library_name: Distily
tags:
  - Distily

Distillation dataset created with Distily.

Method: Generated sequences randomly with temperature config 1.0
Model URI: gpt2
Number of Samples: 100000
Maximum Sequence Length: 1024 tokens