Banana

Public Access GPTJ

Use a production ready GPTJ in <1 min. Free, forever.

Add this to your python code:

import banana_dev as banana

api_key={your key here}
model_key="gptj"
model_parameters = { "text": "Hey GPTJ! How are you?", "length": 50, "temperature": 0.9, "topK": 50, "topP": 0.9}

out = banana.run(api_key, model_key, model_parameters)
print(out)

Arguments:

Arg Description Required Type Example
api_key Your API key, found on the User Dashboard True string "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx'
model_key This models name True string "gptj"
model_parameters Dictionary of custom tuning parameters False dict { "text": "Hey GPTJ! How are you?", "length": 50, "temperature": 0.9, "topK": 50, "topP": 0.9}

Fine-Tuning:

Adjusting these parameters from the default is not recommended.

Arg Description Required Type Example
text The input text to prompt generation True string Hey GPTJ! How are you?
temperature A value between 0.1 and 1, to adjust randomness.

- Smaller values create seemingly random output.
- Larger values create repeating phrases in the output.
False (default=0.8) float 0.8
length How many tokens to generate. False (default=50)) int 250
topP A value between 0.1 and 1, to adjust randomness.

- Smaller values create seemingly random output.
- Larger values create repeating phrases in the output.
False (default=0.8) float 0.8
topK an integer >=1. Top k sampling means sorts potential output generations by probability and returns only the topK likely outputs, the rest are zero'd out. False float 50