INDEX

Explanations

references to locker rooms

oai_token-act-pair · gpt-3.5-turbo

mentions of locker rooms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

hillary

-0.77

PsyNetMessage

-0.70

alez

-0.69

Duration

-0.66

ulhu

-0.65

romy

-0.64

ical

-0.64

 Cause

-0.61

EFF

-0.61

nom

-0.61

POSITIVE LOGITS

 locker

1.12

 room

1.05

 rooms

1.04

room

0.97

 Room

0.89

 drawer

0.85

 closet

0.83

lain

0.81

pees

0.81

robe

0.79

Activations Density 0.040%

references to locker rooms

mentions of locker rooms

No Comments

No Known Activations