INDEX

Explanations

programming language constructs and keywords related to statements and expressions

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

small-caps and underscore prefixes

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

KommentareTeilen

-0.49

yarnpkg

-0.48

 surla

-0.46

 defaultstate

-0.44

openConnection

-0.40

 mistrzost

-0.40

ActionCreators

-0.40

 تضيفلها

-0.38

almente

-0.38

icoot

-0.38

POSITIVE LOGITS

PerformLayout

0.48

-------------</

0.48

RegressionTest

0.48

Билгалдахарш

0.45

Історія

0.44

 dAtA

0.43

 propOrder

0.42

invokeLater

0.42

verifyException

0.42

Demografia

0.41

Activations Density 0.017%

programming language constructs and keywords related to statements and expressions

small-caps and underscore prefixes

No Comments

No Known Activations