INDEX

Explanations

phrases related to political leadership roles

oai_token-act-pair · gpt-3.5-turbo

references to political leadership and majority positions in Congress

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

adr

-0.87

worldly

-0.71

 worldly

-0.69

blem

-0.66

iverpool

-0.66

iceps

-0.65

bsite

-0.64

spr

-0.64

ikh

-0.64

acl

-0.61

POSITIVE LOGITS

 Majority

1.07

 Whip

0.92

 Leader

0.86

OTUS

0.84

 Votes

0.80

 majorities

0.80

majority

0.80

rats

0.72

ority

0.70

 Pelosi

0.70

Activations Density 0.008%

phrases related to political leadership roles

references to political leadership and majority positions in Congress

No Comments

No Known Activations