INDEX

Explanations

phrases related to political and legal activities

oai_token-act-pair · gpt-3.5-turbo

phrases or structures indicating details about subjects or topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

plet

-0.74

ority

-0.73

orrow

-0.73

encies

-0.71

 millenn

-0.71

 occas

-0.70

anches

-0.69

 intimidated

-0.69

 Palestin

-0.69

ividual

-0.69

POSITIVE LOGITS

TR

0.73

VAL

0.70

ãĢİ

0.69

Hash

0.68

ARS

0.67

Ma

0.67

ERO

0.67

VIS

0.67

UE

0.66

DX

0.66

Activations Density 0.425%

phrases related to political and legal activities

phrases or structures indicating details about subjects or topics

No Comments

No Known Activations