INDEX

Explanations

mentions of the abbreviation "SM."

oai_token-act-pair · gpt-3.5-turbo

references to social media

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

Ved

-0.71

 Frie

-0.68

reon

-0.68

 Authorization

-0.68

ãĥĥãĥĪ

-0.67

 Gamergate

-0.67

icular

-0.66

bart

-0.66

ãĤ¡

-0.64

Ã§a

-0.64

POSITIVE LOGITS

ITH

1.11

ART

1.03

ugg

0.98

iley

0.98

ooth

0.92

oke

0.91

ASH

0.90

achine

0.87

ORPG

0.85

ores

0.82

Activations Density 0.013%

mentions of the abbreviation "SM."

references to social media

No Comments

No Known Activations