INDEX

Explanations

phrases related to covert operations or espionage

oai_token-act-pair · gpt-3.5-turbo

terms related to secretive or hidden operations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ð

-0.77

hner

-0.71

Fam

-0.70

APH

-0.69

ulz

-0.68

 Atkinson

-0.68

ICAN

-0.68

ĪĴ

-0.68

Interstitial

-0.65

ordan

-0.65

POSITIVE LOGITS

 covert

1.07

entimes

0.92

 infiltr

0.87

ly

0.83

 operatives

0.82

 plot

0.81

 espionage

0.80

clo

0.77

CIA

0.75

 spying

0.75

Activations Density 0.007%

phrases related to covert operations or espionage

terms related to secretive or hidden operations

No Comments

No Known Activations