INDEX

Explanations

words related to names, especially those starting with 'Whit'

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Whit."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

PsyNetMessage

-0.85

 GEAR

-0.80

igslist

-0.77

aeda

-0.74

raltar

-0.73

orsi

-0.72

anwhile

-0.72

Pg

-0.68

DIR

-0.67

VID

-0.66

POSITIVE LOGITS

ewater

1.34

eness

1.14

ening

1.06

bread

1.06

ened

1.01

eless

1.01

elist

0.97

estone

0.95

acre

0.94

ework

0.94

Activations Density 0.016%

words related to names, especially those starting with 'Whit'

mentions of the name "Whit."

No Comments

No Known Activations