INDEX

Explanations

government and political officials

oai_token-act-pair · gpt-3.5-turbo

references to government and military officials

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

Hig

-0.72

yss

-0.71

Mech

-0.71

à¨

-0.69

Beg

-0.69

ãĥ¥

-0.66

à¤

-0.65

ãĥĥ

-0.64

OPER

-0.63

ãĥ¡

-0.63

POSITIVE LOGITS

ervatives

0.90

hops

0.90

paces

0.84

 stationed

0.84

hips

0.79

 alike

0.76

hip

0.70

 tasked

0.69

enei

0.68

who

0.67

Activations Density 0.250%

government and political officials

references to government and military officials

No Comments

No Known Activations