Spaces:

nicholasKluge
/

Aira-Demo

Running

nicholasKluge commited on Jun 18, 2023

Commit

b31de11

•

1 Parent(s): 5e3a56d

Update app.py

Files changed (1) hide show

app.py CHANGED Viewed

@@ -3,7 +3,7 @@ import torch
 import gradio as gr
 from transformers import AutoTokenizer, AutoModelForCausalLM
-model_id = "nicholasKluge/Aira-Instruct-124"
 token = "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"
 device = "cuda" if torch.cuda.is_available() else "cpu"
@@ -27,7 +27,7 @@ We developed our open-domain conversational chatbots via conditional text genera
 🤥 Generative models can perpetuate the generation of pseudo-informative content, that is, false information that may appear truthful.
-🤬 In certain types of tasks, generative models can produce harmful and discriminatory content inspired by historical stereotypes against sensitive attributes (for example, gender, race, and religion).
 `Aira` is intended only for academic research. For more information, visit our [HuggingFace models](https://huggingface.co/nicholasKluge) to see how we developed `Aira`.
 """

 import gradio as gr
 from transformers import AutoTokenizer, AutoModelForCausalLM
+model_id = "nicholasKluge/Aira-Instruct-124M"
 token = "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"
 device = "cuda" if torch.cuda.is_available() else "cpu"
 🤥 Generative models can perpetuate the generation of pseudo-informative content, that is, false information that may appear truthful.
+🤬 In certain types of tasks, generative models can produce harmful and discriminatory content inspired by historical stereotypes.
 `Aira` is intended only for academic research. For more information, visit our [HuggingFace models](https://huggingface.co/nicholasKluge) to see how we developed `Aira`.
 """