INDEX

Explanations

phrases indicating knowledge or awareness of something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "know" 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 @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

phrine

-0.82

oples

-0.82

ItemTracker

-0.81

interstitial

-0.79

otion

-0.76

conservancy

-0.73

pite

-0.71

 Yugoslavia

-0.71

atism

-0.70

aredevil

-0.69

POSITIVE LOGITS

ledge

1.14

ledged

1.06

lege

1.02

LED

0.91

 beforehand

0.77

ariat

0.76

how

0.76

edge

0.76

hent

0.72

how

0.70

Activations Density 0.064%

phrases indicating knowledge or awareness of something

instances of the word "know" in various contexts

No Comments

No Known Activations