INDEX

Explanations

punctuation marks, specifically commas

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

programming terms and keywords

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

发表于

-0.53

makeText

-0.52

UrlEncoded

-0.51

inSlope

-0.46

 وتسجيلات

-0.45

+#+#

-0.43

ceğine

-0.42

Anytime

-0.41

msgstr

-0.40

 wireType

-0.40

POSITIVE LOGITS

SupportActionBar

0.38

 Signalez

0.38

रोध

0.37

 bénévol

0.36

例の

0.35

CppMethod

0.35

 Федерации

0.35

Obrázky

0.34

astéro

0.33

 secours

0.33

Activations Density 0.000%

punctuation marks, specifically commas

programming terms and keywords

No Comments

No Known Activations