INDEX

Explanations

references to hacking activities and related terms, such as software, emails, and cyber attacks

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ufc

-0.55

chemy

-0.51

 momo

-0.50

AndroidJUnit

-0.50

 polenta

-0.50

 Woodruff

-0.49

comfor

-0.48

 Sumter

-0.48

Miejs

-0.47

 Beno

-0.47

POSITIVE LOGITS

 hack

1.35

hack

1.25

 Hack

1.24

Hack

1.22

 hacking

1.21

 hacks

1.11

 hacked

1.08

 Hacking

1.05

 hackers

1.03

 HACK

1.00

Activations Density 0.070%

references to hacking activities and related terms, such as software, emails, and cyber attacks

No Comments

No Known Activations