INDEX

Explanations

names of people and places

oai_token-act-pair · gpt-3.5-turbo

numerical data and significant cultural or historical references

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 crunch

-0.81

 princ

-0.80

 casc

-0.79

 tiss

-0.78

 tightening

-0.75

 eleph

-0.73

 answ

-0.73

 neighb

-0.71

 shoppers

-0.69

llah

-0.68

POSITIVE LOGITS

References

1.56

Appearances

1.46

Trivia

1.43

Notable

1.34

Gallery

1.34

History

1.31

See

1.30

Contents

1.30

Other

1.27

During

1.26

Activations Density 0.218%

names of people and places

numerical data and significant cultural or historical references

No Comments

No Known Activations