INDEX

Explanations

function calls and output statements in code

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

looping or iterating over items

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

TagNumber

-0.56

RTSN

-0.55

UnknownFieldSet

-0.54

 Normdatei

-0.54

principalColumn

-0.53

 otomatig

-0.52

Datuak

-0.52

 getItemCount

-0.52

oneofs

-0.50

cookieParser

-0.50

POSITIVE LOGITS

ParallelGroup

0.50

hom

0.41

 individual

0.40

each

0.40

 each

0.39

 propositions

0.37

 suggestions

0.36

 Kere

0.36

clo

0.35

 individuel

0.35

Activations Density 0.005%

function calls and output statements in code

looping or iterating over items

No Comments

No Known Activations