YAML Metadata Error:"value" must be of type object
Configuration Parsing Warning:Config file config.json cannot be fetched (too big)
Configuration Parsing Warning:Config file tokenizer_config.json cannot be fetched (too big)
Petrained Model Description: Open Source Version of GPT-3
Generative Pre-trained Transformer 3 (GPT-3) is an autoregressive language model that uses deep learning to produce human-like text. It is the third-generation language prediction model in the GPT-n series (and the successor to GPT-2) created by OpenAI
GPT-Neo (125M) is a transformer model designed using EleutherAI's replication of the GPT-3 architecture. GPT-Neo refers to the class of models, while 125M represents the number of parameters of this particular pre-trained model. and first released in this repository.
Fine-tuned Model Description: GPT-3 fine-tuned Multi-XScience
The Open Source version of GPT-3: GPT-Neo(125M) has been fine-tuned on a dataset called "Multi-XScience": Multi-XScience_Repository: A Large-scale Dataset for Extreme Multi-document Summarization of Scientific Articles.
I first fine-tuned and then deployed it using Google "Material Design" (on Anvil): Abir Scientific text Generator
By fine-tuning GPT-Neo(Open Source version of GPT-3), on Multi-XScience dataset, the model is now able to generate scientific texts(even better than GPT-J(6B).
Try putting the prompt "attention is all" on both my Abir Scientific text Generator and on the GPT-J Eleuther.ai Demo to understand what I mean.
And Here's a demonstration video for this. Video real-time Demontration
- Downloads last month
- 11