INDEX

Explanations

proper nouns and specific references

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

Japanese text or language

np_acts-logits-general · gemini-2.0-flash

そいつ

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

kijken

-0.59

ädie

-0.57

 عاما

-0.56

 Larson

-0.54

 divisão

-0.54

 lccn

-0.53

 Pollutant

-0.52

garmin

-0.52

ussia

-0.52

್ದ

-0.51

POSITIVE LOGITS

NameInMap

0.57

 linkovi

0.57

ScopeManager

0.53

TagHelper

0.52

ISPR

0.52

]--;

0.49

 htons

0.47

Bronnen

0.47

Diweddarwch

0.44

mappedBy

0.42

Activations Density 0.045%

proper nouns and specific references

Japanese text or language

そいつ

No Comments

No Known Activations