INDEX

Explanations

names of specific individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

atform

-0.96

sburgh

-0.94

icular

-0.87

ures

-0.87

isphere

-0.87

ional

-0.84

ogene

-0.83

nostic

-0.83

lessly

-0.83

icles

-0.83

POSITIVE LOGITS

 Rowe

1.27

Row

0.76

 FANTASY

0.75

 Christy

0.73

OWN

0.73

idge

0.71

 Russo

0.71

 Kitty

0.70

dden

0.69

VEL

0.69

Activations Density 0.028%

names of specific individuals

proper nouns, particularly names of individuals

No Comments

No Known Activations