INDEX

Explanations

words related to political figures and events

oai_token-act-pair · gpt-3.5-turbo

phrases that suggest a legal or authoritative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 myster

-0.65

Berry

-0.52

 STUD

-0.52

Owl

-0.51

 Revival

-0.51

CLASSIFIED

-0.48

 Collider

-0.48

MAP

-0.47

 WEEK

-0.47

 Palestin

-0.47

POSITIVE LOGITS

tarian

0.72

own

0.72

emic

0.69

oker

0.69

etary

0.68

audi

0.68

ega

0.68

oes

0.67

ember

0.66

daq

0.66

Activations Density 0.450%

words related to political figures and events

phrases that suggest a legal or authoritative context

No Comments

No Known Activations