INDEX

Explanations

political affiliations or ideologies

oai_token-act-pair · gpt-3.5-turbo

terms related to political affiliations and ideologies

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

hod

-0.73

kens

-0.73

enge

-0.73

 Haku

-0.72

ysis

-0.72

 Kard

-0.69

usable

-0.68

Ð

-0.68

thood

-0.67

urable

-0.65

POSITIVE LOGITS

 bloc

0.96

 leaning

0.95

 powerhouse

0.89

 Democrats

0.88

 caucus

0.86

 voters

0.86

 strongh

0.85

 voices

0.85

Democrats

0.85

 stronghold

0.85

Activations Density 0.074%

political affiliations or ideologies

terms related to political affiliations and ideologies

No Comments

No Known Activations