INDEX

Explanations

references to espionage and spy-related content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

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

xual

-0.91

esville

-0.80

à¼

-0.74

clus

-0.72

ĸļ

-0.71

 Kurd

-0.66

Interstitial

-0.65

 Explain

-0.65

kell

-0.64

urther

-0.63

POSITIVE LOGITS

glass

1.00

oleon

0.93

sonian

0.83

ware

0.83

dropping

0.82

 satellites

0.82

OSH

0.79

ionage

0.79

 spying

0.77

agency

0.76

Activations Density 0.026%

references to espionage and spy-related content

No Comments

No Known Activations