INDEX

Explanations

words indicating importance or emphasis in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

drop

-0.45

 drop

-0.45

strtolower

-0.44

 drops

-0.43

-0.42

Brien

-0.42

Big

-0.41

printLine

-0.41

 Blue

-0.41

him

-0.39

POSITIVE LOGITS

tvguidetime

1.18

RegistryLite

1.00

 CreateTagHelper

0.95

SerializedSize

0.89

 doInBackground

0.89

UnsafeEnabled

0.87

 disambiguazione

0.86

verwijspagina

0.84

RegressionTest

0.84

Personensuche

0.83

Activations Density 0.655%

words indicating importance or emphasis in various contexts

No Comments

No Known Activations