INDEX

Explanations

specific formatting elements or code-related structures

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

rounding to dp

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

SourceChecksum

-0.47

 Eisenberg

-0.46

 labour

-0.45

 Larkin

-0.45

Bom

-0.44

 namorado

-0.42

 esfera

-0.42

ServletContext

-0.40

โคร

-0.40

flink

-0.39

POSITIVE LOGITS

NameInMap

0.61

MongoClient

0.52

 שוליים

0.51

XmlEnum

0.49

الدراسه

0.48

 propOrder

0.47

دانشنامهٔ

0.47

ंदीखरीदारी

0.45

hbs

0.45

isEmail

0.43

Activations Density 0.000%

specific formatting elements or code-related structures

rounding to dp

No Comments

No Known Activations

specific formatting elements or code-related structures

rounding to dp

No Comments

No Known Activations