INDEX

Explanations

indicators of statistical significance and results

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

foreign words

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

 Chwiliwch

-0.40

IVEREF

-0.40

 symbolically

-0.40

RTLD

-0.38

queryInterface

-0.37

Життєпис

-0.37

 grom

-0.37

 Mesopotamia

-0.37

🇭

-0.36

 Freire

-0.36

POSITIVE LOGITS

 insufic

0.48

還好

0.45

 tranquilidad

0.43

 zelve

0.42

ATILE

0.41

ьаж

0.41

ostante

0.40

ukkig

0.40

brata

0.39

 تانيه

0.39

Activations Density 0.192%

indicators of statistical significance and results

foreign words

No Comments

No Known Activations