INDEX

Explanations

terms related to political events and controversies

oai_token-act-pair · gpt-3.5-turbo

occurrences of sensitive political information and events

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gren

-0.66

course

-0.66

boro

-0.65

cius

-0.65

gements

-0.64

lihood

-0.64

Cat

-0.62

utical

-0.62

 manoeuv

-0.61

iates

-0.60

POSITIVE LOGITS

ÂłÂłÂł

0.74

arta

0.71

ÂłÂłÂłÂłÂłÂłÂłÂł

0.70

Specifically

0.69

ÂłÂłÂłÂł

0.65

ÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂł

0.65

WASHINGTON

0.64

Writing

0.63

 WARN

0.63

 REUTERS

0.61

Activations Density 0.331%

terms related to political events and controversies

occurrences of sensitive political information and events

No Comments

No Known Activations