INDEX

Explanations

concepts related to data security and the protection of sensitive information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

sti

-0.38

upol

-0.36

build

-0.35

Fix

-0.34

jspx

-0.34

FAS

-0.34

FAS

-0.34

Sim

-0.33

lif

-0.33

Strain

-0.33

POSITIVE LOGITS

 privacy

0.68

 Privacy

0.62

 security

0.62

 PRIVACY

0.60

Privacy

0.60

privacy

0.59

 confidentiality

0.59

 referrerpolicy

0.56

 secure

0.54

 sécur

0.52

Activations Density 0.334%

concepts related to data security and the protection of sensitive information

No Comments

No Known Activations