INDEX

Explanations

repeated graphical or dividers in formatted content

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

genetic sequences

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

WebServlet

-0.58

<h2>

-0.55

 getCode

-0.52

"=

-0.52

 sqlSession

-0.52

"*.

-0.50

 zrozum

-0.50

 UnityEngine

-0.49

«<

-0.47

 getInt

-0.47

POSITIVE LOGITS

<h5>

1.10

</h5>

0.66

standers

0.56

 Geophys

0.54

#####

0.53

ecko

0.52

 #####

0.51

épaisseur

0.50

stander

0.50

ள்

0.49

Activations Density 0.012%

repeated graphical or dividers in formatted content

genetic sequences

No Comments

No Known Activations