Edit model card

GPT-2 NOS INSTAGRAM ANALYZER AI

examples

Nos AI is an artificial intelligence software that utilizes the GPT-2 language model to analyze Instagram users in real-time at specified intervals and communicates this information to the user. For instance, Nos attempts to find out whom an Instagram user most recently sent a request to or where a specific account's name has appeared, either in posts or biographies, by examining the accounts followed by the target user. All these operations are conducted by leveraging Instagram APIs. To use the program, Nos needs to log in to any account.

Model Description

The GPT-2 model used by Nos was trained on a small dataset consisting of specific actions. The current GPT-2 Model has 124 million hyperparameters. It generates text based on users' actions on their accounts. Nos GPT2 first collects the accounts followed by the target account. During the initial collection, it perceives them as newly followed accounts and records the initially checked users. Later on, it checks whether the analyzed user's name appears in the accounts they follow. For this purpose, it focuses on the writings, comments, and biographies of the accounts followed by the analyzed user. If the analyzed user's name is found among the followed accounts, Nos promptly informs you. Subsequently, it attempts to collect the open-source data of the analyzed user from the internet, completing the initial check this way. In the next analysis, if the analyzed user has followed someone new within the specified time frame, Nos directly informs you about this person and analyzes all their information.

  • Check out the github page for the program

Training

  • This GPT-2 model is rated for an average of 1 hour with an A100 GPU.

Authors

Downloads last month
15
Safetensors
Model size
124M params
Tensor type
F32
·
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.