INDEX

Explanations

quantitative data and numerical values in the document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

RegressionTest

-0.45

 espirituales

-0.41

velmente

-0.40

Lähteet

-0.39

نیم

-0.38

 problemática

-0.36

Viitteet

-0.36

 mesmas

-0.36

Życiorys

-0.36

nictwa

-0.35

POSITIVE LOGITS

stateProvider

0.56

Largest

0.54

 vast

0.53

featureID

0.52

 largest

0.52

 average

0.51

 gigantic

0.51

largest

0.50

 Largest

0.50

Activations Density 0.252%

quantitative data and numerical values in the document

No Comments

No Known Activations