INDEX

Explanations

references to specific locations and organizational contexts

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

exitRule

-0.38

 filling

-0.37

fill

-0.36

WebElementEntity

-0.33

 Filling

-0.33

 thanks

-0.32

 anything

-0.32

 fill

-0.32

 AssemblyCulture

-0.31

 checking

-0.31

POSITIVE LOGITS

 betweenstory

0.63

 userEmail

0.55

TryDecodeAsNil

0.54

$_(

0.53

BeginContext

0.52

 Vikipedi

0.52

 GetYourGuide

0.52

llavero

0.52

twimg

0.49

 getCity

0.49

Activations Density 0.073%

references to specific locations and organizational contexts

No Comments

No Known Activations