INDEX

Explanations

terms and phrases related to proving mathematical or logical statements

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

in this case we have

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

RTLR

-0.59

 الرياضيه

-0.56

AddTagHelper

-0.56

 Meksiku

-0.55

Tembelea

-0.54

 defaultstate

-0.52

 незавершена

-0.50

StoryboardSegue

-0.47

 unknownFields

-0.47

 JpaRepository

-0.46

POSITIVE LOGITS

 ones

0.41

orteur

0.39

athread

0.38

plet

0.37

gsub

0.36

зья

0.36

 Schell

0.36

cocho

0.36

meen

0.36

 AGRE

0.36

Activations Density 0.019%

terms and phrases related to proving mathematical or logical statements

in this case we have

No Comments

No Known Activations