INDEX

Explanations

terms related to the concept of "mal" or "badness," particularly in relation to individuals and actions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

"Mal" followed by other letters

np_acts-logits-general · gemini-2.0-flash

Malala and similar words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 createState

-0.73

++]=

-0.55

curities

-0.55

 trouvait

-0.54

 Clermont

-0.53

 signé

-0.52

 préfé

-0.50

alyptus

-0.50

 FontWeight

-0.50

={({

-0.49

POSITIVE LOGITS

Mal

0.85

Mal

0.84

MAL

0.81

MAL

0.72

tvguidetime

0.72

+:+

0.68

mal

0.67

 EconPapers

0.64

mal

0.60

richTextPanel

0.59

Activations Density 0.099%

terms related to the concept of "mal" or "badness," particularly in relation to individuals and actions

"Mal" followed by other letters

Malala and similar words

No Known Activations