INDEX

Explanations

specific dates, locations, and historical references

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

geographic and temporal entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 galore

-0.53

Personendaten

-0.53

ientôt

-0.50

 disambiguazione

-0.49

 besz

-0.49

 Rein

-0.49

Diweddarwch

-0.47

alities

-0.45

 приме

-0.45

atisk

-0.45

POSITIVE LOGITS

 snippetHide

0.55

ConstraintMaker

0.46

 Wicidata

0.43

adpleegd

0.43

 dAtA

0.42

wapV

0.37

 conven

0.36

EDEFAULT

0.36

 ویکی‌پدی

0.35

WriteAttribute

0.35

Activations Density 0.062%

specific dates, locations, and historical references

geographic and temporal entities

No Comments

No Known Activations