INDEX

Explanations

phrases or words related to political moderation

oai_token-act-pair · gpt-3.5-turbo

the term "moderate" in various contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

arium

-0.88

chy

-0.74

Report

-0.73

STON

-0.72

AMES

-0.72

æ©

-0.71

yang

-0.70

ADRA

-0.70

raltar

-0.70

ynthesis

-0.69

POSITIVE LOGITS

erate

0.97

 minded

0.87

 sized

0.85

 medi

0.84

 easing

0.76

 leaning

0.71

 moderate

0.70

lihood

0.69

 fare

0.68

xual

0.68

Activations Density 0.015%

phrases or words related to political moderation

the term "moderate" in various contexts

No Comments

No Known Activations