INDEX

Explanations

terms related to political advisors and strategists

oai_token-act-pair · gpt-3.5-turbo

names of political figures and strategists

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

phis

-0.89

ivity

-0.80

 SERVICE

-0.75

 Jagu

-0.74

EVA

-0.73

 Gael

-0.71

 Ranked

-0.70

venge

-0.69

stice

-0.67

angular

-0.67

POSITIVE LOGITS

 Bannon

0.94

hattan

0.79

fried

0.79

espie

0.79

 aides

0.79

glers

0.78

 confid

0.77

 allies

0.75

olini

0.74

umenthal

0.74

Activations Density 0.012%

terms related to political advisors and strategists

names of political figures and strategists

No Comments

No Known Activations