INDEX

Explanations

terms associated with accusations or claims of wrongdoing

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

unproven accusations

np_acts-logits-general · gemini-2.0-flash

alleged events or issues

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

KommentareTeilen

-0.70

abetes

-0.54

 BoxDecoration

-0.50

 BoxFit

-0.47

 OFDb

-0.46

 betweenstory

-0.46

-------

-0.45

asjons

-0.44

engel

-0.44

Kanpo

-0.43

POSITIVE LOGITS

 alleged

1.01

 supposed

0.94

 purported

0.94

 supposedly

0.82

 allegedly

0.82

supposed

0.80

 putative

0.79

 supuestamente

0.78

 presumed

0.73

 prétend

0.73

Activations Density 0.457%

terms associated with accusations or claims of wrongdoing

unproven accusations

alleged events or issues

No Comments

No Known Activations