INDEX

Explanations

phrases related to covert operations and clandestine activities

oai_token-act-pair · gpt-3.5-turbo

terms related to covert operations and secrecy

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

-0.82

ĪĴ

-0.81

APH

-0.73

ordan

-0.73

ICAN

-0.72

Interstitial

-0.71

alon

-0.70

 Atkinson

-0.70

annis

-0.68

CENT

-0.68

POSITIVE LOGITS

 covert

0.99

entimes

0.83

ops

0.83

ly

0.81

 operatives

0.81

 infiltr

0.78

CIA

0.77

 espionage

0.77

 sneak

0.77

 plot

0.77

Activations Density 0.013%

phrases related to covert operations and clandestine activities

terms related to covert operations and secrecy

No Comments

No Known Activations