INDEX

Explanations

phrases related to intelligence agencies and officials

oai_token-act-pair · gpt-3.5-turbo

references to national intelligence and intelligence agencies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

loading

-0.83

anan

-0.76

LOAD

-0.75

ged

-0.73

zl

-0.72

unin

-0.71

joy

-0.71

interstitial

-0.71

shr

-0.70

¶ħ

-0.69

POSITIVE LOGITS

 Agency

0.99

 Intelligence

0.87

 Surveillance

0.85

 Oversight

0.82

elligence

0.80

 Accountability

0.79

 Labs

0.79

 Laboratory

0.79

 Observatory

0.79

 Bureau

0.78

Activations Density 0.019%

phrases related to intelligence agencies and officials

references to national intelligence and intelligence agencies

No Comments

No Known Activations