INDEX

Explanations

programming-related terms and operations

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

path separators

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

rrggbb

-0.47

 SqlCommand

-0.40

+#+#

-0.36

Occur

-0.35

 SqlParameter

-0.35

 prophetic

-0.34

 biß

-0.34

 ashtray

-0.34

}$

-0.33

 jambe

-0.33

POSITIVE LOGITS

 kasarigan

0.56

 ویکی‌پدی

0.55

GEBURTSDATUM

0.54

 հղումներ

0.54

出版年

0.49

 uniformes

0.47

جستارهای

0.47

__":

0.45

RTSN

0.45

合同

0.43

Activations Density 0.009%

programming-related terms and operations

path separators

No Comments

No Known Activations