INDEX

Explanations

instances of the word "see" or its variations

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

see in context

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

the

-0.53

 Paglinawan

-0.50

räck

-0.46

Normdaten

-0.45

ortecedor

-0.44

 própri

-0.44

 trilha

-0.44

odziel

-0.44

تقاوى

-0.43

 spørs

-0.42

POSITIVE LOGITS

in

0.56

resultCode

0.52

 elsewhere

0.50

extAlignment

0.49

 here

0.48

 around

0.48

__((

0.48

SaveChanges

0.48

 packageName

0.47

 MyClass

0.47

Activations Density 0.015%

instances of the word "see" or its variations

see in context

No Comments

No Known Activations