INDEX

Explanations

proper nouns related to politics and individuals in the context of investigations

oai_token-act-pair · gpt-3.5-turbo

mentions of specific names or figures, particularly political ones

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

ERAL

-0.81

cation

-0.69

DCS

-0.69

hered

-0.68

hetto

-0.68

CoC

-0.67

iary

-0.67

reet

-0.64

ician

-0.62

imates

-0.62

POSITIVE LOGITS

xus

0.91

 Rousse

0.79

yss

0.75

bra

0.73

stal

0.72

lete

0.70

acl

0.69

thia

0.69

lda

0.68

rl

0.68

Activations Density 0.031%

proper nouns related to politics and individuals in the context of investigations

mentions of specific names or figures, particularly political ones

No Comments

No Known Activations