INDEX

Explanations

instances where the phrase "of course" is used

oai_token-act-pair · gpt-3.5-turbo

phrases that express certainty or acknowledgment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Dialog

-0.81

Hide

-0.75

tex

-0.74

crow

-0.73

HUD

-0.72

panel

-0.71

cit

-0.69

idas

-0.69

vert

-0.68

Released

-0.67

POSITIVE LOGITS

 caveats

0.88

 caveat

0.79

 none

0.69

 exceptions

0.68

 plenty

0.68

 begs

0.68

 never

0.67

 wouldn

0.66

 assumes

0.63

 surprises

0.63

Activations Density 0.138%

instances where the phrase "of course" is used

phrases that express certainty or acknowledgment

No Comments

No Known Activations