Select your cookie preferences

We use essential cookies and similar tools that are necessary to provide our site and services. We use performance cookies to collect anonymous statistics, so we can understand how customers use our site and make improvements. Essential cookies cannot be deactivated, but you can choose “Customize” or “Decline” to decline performance cookies.

If you agree, AWS and approved third parties will also use cookies to provide useful site features, remember your preferences, and display relevant content, including relevant advertising. To accept or decline all non-essential cookies, choose “Accept” or “Decline.” To make more detailed choices, choose “Customize.”
Sign in
Sign in
or
Create a new account
About
Categories
Delivery Methods
Solutions
Resources
 Your Saved List Become a Channel Partner Sell in AWS Marketplace Amazon Web Services Home Help

Amazon Sagemaker

Amazon SageMaker is a fully-managed platform that enables developers and data scientists to quickly and easily build, train, and deploy machine learning models at any scale. With Amazon SageMaker, all the barriers and complexity that typically slow down developers who want to use machine learning are removed. The service includes models that can be used together or independently to build, train, and deploy your machine learning models.

product logo

GPT-J 6B (GPT-3) | Text Generation

By:
not provided
The most powerful text generation AI available on Sagemaker
  • This product has been removed and is no longer available to new customers.

Product Overview

Access GPT-J 6B for text generation with a simple API. You can simply provide text or customize many of the parameters in a json format to architect a prompt for the AI. The AI will generate text in response. API documentation is provided at https://aixsolutionsgroup.com/aws-sagemaker-support . GPT-J 6b is currently the largest and most advanced open-source GPT engine. We also offer all versions of the GPT Neo series and GPT-2 series: GPT Neo 125M, GPT Neo 1.3B, GPT Neo 2.7B, GPT J 6B, GPT-2 Tiny, GPT-2 Small, GPT-2 Medium, GPT-2 Large, GPT-2 XL. GPT-J 6B (6B parameters compared to GPT2 XL's 1.6B) is the most powerful text generation engine available on AWS Sagemaker Marketplace, with performance comparable to OpenAI's GPT-3's Curie engine, the second most powerful OpenAI GPT-3 engine (with DaVinci being the first).

Key Data

By
not provided
Categories
Generation-Text
Text
Natural Language Processing
Type
Model Package
Fulfillment Methods
Amazon SageMaker

Highlights

  • The most powerful text generation AI available on Sagemaker, with performance comparable to OpenAI GPT-3 Curie engine.

Not quite sure what you’re looking for? AWS Marketplace can help you find the right solution for your use case. Contact us

Usage Information

Model input and output details

Input

Summary

The model accepts either json or plain text.

If given plain text, the entirety of the plain text will be assumed to be text input for the model, and default values for the other parameters will be used.

If given json, the json will be parsed and checked for the available parameters.

Input MIME type
application/x-www-form-urlencoded, application/json
Sample input data
Expand all input descriptions
Field name
input_prompt
Description
The prompt for the generative model.
Required
No
Data type
FreeText
Default value
Minimum length: 1 character. Maximum length: 8192 characters
Field name
token_min_length
Description
Minimum length of tokens to generate. One token equals about 4 characters of text. If token_max_length provided, token_min_length must be less than or equal to token_max_length. This value is ignored if the stop sequence occurs first.
Required
No
Data type
Integer
Range
Min: 1
Max: 2048
Default value
25
Field name
token_max_length
Description
Maximum length of tokens to generate. One token equals about 4 characters of text. If token_min_length provided, token_max_length must be greater than token_min_length. This value is ignored if the stop sequence occurs first.
Required
No
Data type
Integer
Range
Min: 1
Max: 2048
Default value
25
Field name
temperature
Description
Controls randomness. Lowering results in less random completions. As the temperature approaches zero, the model will become deterministic and repetitive.
Required
No
Data type
Continuous (Float)
Range
Min: 0
Max: 1
Default value
1.0
Field name
top_p
Description
Controls diversity via nucleus sampling: 0.5 means half of all likelihood-weighted options are considered.
Required
No
Data type
Continuous (Float)
Range
Min: 0
Max: 1
Default value
0.9
Field name
top_k
Description
Controls the number of highest probability vocabulary tokens to keep for top-k-filtering.
Required
No
Data type
Integer
Range
Min: 0
Max: 100
Default value
50
Field name
stop_sequence
Description
A sequence where the API will stop generating further tokens. The returned text will not contain the stop sequence.
Required
No
Data type
FreeText
Default value
None
Field name
num_beams
Description
When generating text, beam search reduces the risk of missing hidden high probability word sequences by keeping the most likely num_beams of hypotheses at each time step and eventually choosing the hypothesis that has the overall highest probability.
Required
No
Data type
Integer
Range
Min: 1
Max: 100
Default value
1
Field name
no_repeat_ngram_size
Description
If set to greater than 0, all ngrams of that size can only occur once.
Required
No
Data type
Integer
Range
Min: 0
Max: 100
Default value
100
Field name
early_stopping
Description
Whether to stop the beam search when at least num_beams sentences are finished per batch or not.
Required
No
Data type
Categorical
Default value
false
Allowed values
true,false
Field name
do_sample
Description
Whether or not to use sampling; use greedy decoding otherwise.
Required
No
Data type
Categorical
Default value
false
Allowed values
true,false

Output

Summary

The model outputs the generated text, the compute time, and your input parameters (for your own verification).

Output MIME type
application/json
Sample output data
Expand all output descriptions
Field name
prediction
Description
The prediction by the AI.
Always returned
No
Data type
FreeText
Default value
N/A
Field name
compute_time
Description
The amount of time taken to compute your request.
Always returned
No
Data type
Integer
Range
Default value
N/A
Field name
input_parameters
Description
A dictionary of all of the input parameters used to generate the text. This will be combination of your input and default values, depending on which values you provided.
Always returned
No
Data type
FreeText
Default value
N/A
Field name
error
Description
An error description provided if your call was successfully received but an improper value was provided for a parameter.
Always returned
No
Data type
FreeText
Default value
N/A

Sample notebook

Additional Resources

End User License Agreement

By subscribing to this product you agree to terms and conditions outlined in the product End user License Agreement (EULA)

Support Information

GPT-J 6B (GPT-3) | Text Generation

Call or Text: +1 (415) 888 4745 | Email: info@aixsolutionsgroup.com | Schedule a Meeting: https://calendly.com/josh-wolff-7/aix-chat

AWS Infrastructure

AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.

Learn More

Refund Policy

We do not at this time accept refunds for used resources, but you may cancel your subscription at any time. If you have any concerns, questions, or feedback, please email info@aixsolutionsgroup.com.

Customer Reviews

There are currently no reviews for this product.
View all