INDEX

Explanations

specific names of individuals, likely researchers or authors in academic contexts

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

Followed by capitalized abbreviations or surnames

np_acts-logits-general · gemini-2.0-flash

names with common suffixes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

 totalidad

-0.31

 nabí

-0.29

 médicale

-0.29

 Vogt

-0.28

 alfombra

-0.27

 Verme

-0.27

 zato

-0.27

 druž

-0.26

 provisoire

-0.26

 pravidel

-0.26

POSITIVE LOGITS

iſen

0.83

 myſelf

0.82

ſelben

0.80

 nonUne

0.79

⪜

0.78

 iſt

0.76

iſten

0.75

ſam

0.73

 ſind

0.72

 auffi

0.72

Activations Density 0.572%

specific names of individuals, likely researchers or authors in academic contexts

Followed by capitalized abbreviations or surnames

names with common suffixes

No Comments

No Known Activations