INDEX

Explanations

code segments related to programming syntax, particularly directives and imports

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

code imports and directives

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

 للاسماء

-0.54

 chi̍t

-0.48

jsii

-0.44

 surla

-0.43

VIAF

-0.43

 gynhyrchwyd

-0.41

 cherchés

-0.40

NewUrlParser

-0.39

AutoField

-0.39

BeginInit

-0.38

POSITIVE LOGITS

kolo

0.50

 zwiſchen

0.50

neden

0.48

arang

0.47

mulo

0.47

Steps

0.47

major

0.47

 Register

0.47

 queſto

0.47

itori

0.47

Activations Density 0.001%

code segments related to programming syntax, particularly directives and imports

code imports and directives

No Comments

No Known Activations