INDEX

Explanations

comment block end markers in code

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

code comments and structure

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

iſche

-0.95

ſſung

-0.93

niſſe

-0.92

 ſind

-0.91

 ddelwed

-0.90

ロウィン

-0.89

iſchen

-0.88

$_-

-0.88

[@BOS@]

-0.88

<unused41>

-0.88

POSITIVE LOGITS

 insuffisamment

0.46

0.28

"""

0.27

 Dlatego

0.27

0.26

Hi

0.26

mær

0.26

gesetzt

0.26

0.25

gnings

0.25

Activations Density 0.013%

comment block end markers in code

code comments and structure

No Comments

No Known Activations

comment block end markers in code

code comments and structure

No Comments

No Known Activations