INDEX

Explanations

references to recent violent events or incidents

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

 EconPapers

-0.81

ksesta

-0.50

RTSC

-0.50

TagMode

-0.49

 autorytatywna

-0.48

recated

-0.48

 صوتيه

-0.47

IActionResult

-0.47

IContainer

-0.47

GEBURTSDATUM

-0.45

POSITIVE LOGITS

 lately

0.63

 failures

0.54

 recent

0.54

 scandals

0.52

 recently

0.50

 récentes

0.50

 recente

0.49

 successes

0.47

 announcements

0.47

 occurrences

0.47

Activations Density 0.037%

references to recent violent events or incidents

No Comments

No Known Activations