INDEX

Explanations

phrases and terms indicating consent and user engagement with online services

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

Followed by indefinite pronouns (some, many)

np_acts-logits-general · gemini-2.0-flash

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

 EconPapers

-0.48

یرات

-0.47

Autoritní

-0.45

oa̍t

-0.42

Vidite

-0.40

(&:

-0.39

gnore

-0.37

endal

-0.37

IUrlHelper

-0.36

 legyen

-0.35

POSITIVE LOGITS

 some

2.91

 некоторые

2.48

 algunos

2.48

some

2.45

 некоторых

2.44

 certains

2.42

 niektó

2.39

 alguns

2.36

 algunas

2.34

 niektórych

2.34

Activations Density 1.704%

phrases and terms indicating consent and user engagement with online services

Followed by indefinite pronouns (some, many)

No Comments

No Known Activations