INDEX

Explanations

the prefix "des-" in various words

oai_token-act-pair · gpt-3.5-turbo

words that relate to destruction or degrading

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 laugh

-0.72

ammy

-0.70

mother

-0.67

roach

-0.63

 Izzy

-0.60

wow

-0.60

wered

-0.60

olves

-0.60

obb

-0.60

ELL

-0.60

POSITIVE LOGITS

des

3.74

Des

1.57

des

1.39

Des

1.38

DES

1.32

de

1.31

les

1.14

der

1.04

 undes

1.03

du

1.02

Activations Density 0.014%

the prefix "des-" in various words

words that relate to destruction or degrading

No Comments

No Known Activations