INDEX

Explanations

terms related to guards and protection

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

 utafitiHapana

-0.62

 iscri

-0.61

 verhe

-0.56

 تضيفلها

-0.52

 imparare

-0.51

 espri

-0.49

 chiedere

-0.48

 provare

-0.48

 capire

-0.47

fiore

-0.47

POSITIVE LOGITS

 guard

1.31

 guards

1.22

 Guard

1.18

guard

1.16

 guarding

1.13

Guard

1.10

 Guards

1.10

guards

1.07

 GUARD

1.01

Guards

0.98

Activations Density 0.077%

terms related to guards and protection

No Comments

No Known Activations