INDEX

Explanations

terms related to social and legal issues

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

Follows specific nouns: story, wrap, strike, equation, container, flow, pain, agreement, knock

np_acts-logits-general · gemini-2.0-flash

self-references or specific nouns

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 esetén

-0.73

 Normdatei

-0.65

úgó

-0.58

+#+#

-0.58

TargetException

-0.54

tingly

-0.54

andinavia

-0.54

deauna

-0.54

NameInMap

-0.51

 podjela

-0.50

POSITIVE LOGITS

 itself

1.91

itself

1.44

 Itself

1.33

 itſelf

1.02

本身

0.99

 themselves

0.97

 portion

0.87

 itse

0.84

 zelf

0.82

 himself

0.82

Activations Density 1.462%

terms related to social and legal issues

Follows specific nouns: story, wrap, strike, equation, container, flow, pain, agreement, knock

self-references or specific nouns

No Comments

No Known Activations