INDEX

Explanations

words related to attention, publicity, and coverage

oai_token-act-pair · gpt-3.5-turbo

references to public attention or media coverage

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

externalActionCode

-0.93

NET

-0.76

 partName

-0.73

angelo

-0.70

Ops

-0.66

cking

-0.66

ggies

-0.66

throp

-0.66

*/(

-0.65

--+

-0.65

POSITIVE LOGITS

 announcements

0.65

 nominations

0.64

 candidacy

0.63

 sensational

0.63

 candidates

0.61

 questionable

0.61

 Heath

0.60

effic

0.60

 televised

0.59

 baseless

0.59

Activations Density 0.484%

words related to attention, publicity, and coverage

references to public attention or media coverage

No Comments

No Known Activations