INDEX

Explanations

indicators of statistical manipulation and deception

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

Filmo

-0.44

 gợi

-0.37

Tracce

-0.37

 vertrou

-0.35

 Waray

-0.34

insegna

-0.34

utsche

-0.34

 mixing

-0.33

 APPELLANT

-0.33

 NORTHERN

-0.33

POSITIVE LOGITS

EndGlobalSection

0.54

CloseOperation

0.51

adaptiveStyles

0.51

RenderAtEndOf

0.44

FunctionFlags

0.43

 referenties

0.43

tanleria

0.41

 statistic

0.41

 statistically

0.41

şört

0.40

Activations Density 0.081%

indicators of statistical manipulation and deception

No Comments

No Known Activations