INDEX

Explanations

phrases related to specific chemicals or substances

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "meth" or its variations 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 @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 Clement

-0.78

edom

-0.74

xual

-0.66

yrinth

-0.61

enhagen

-0.59

edo

-0.58

eers

-0.58

Vaj

-0.57

venants

-0.56

éĹĺ

-0.56

POSITIVE LOGITS

lehem

1.51

ulhu

1.15

rower

1.06

ought

1.06

urst

1.06

amphetamine

1.06

iop

1.02

anasia

0.99

anol

0.99

rough

0.99

Activations Density 0.072%

phrases related to specific chemicals or substances

occurrences of the word "meth" or its variations in various contexts

No Known Activations