INDEX

Explanations

names of individuals

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

first names

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

httphttps

-0.59

 onPostExecute

-0.59

évaluateur

-0.56

+#+#

-0.54

 contextLoads

-0.51

Хьажоргаш

-0.50

SharedCtor

-0.50

 Biôgrafia

-0.49

 invokingState

-0.46

qrstuvwxyz

-0.46

POSITIVE LOGITS

 المعيارى

0.53

RTEX

0.44

and

0.40

 parents

0.39

 nakalista

0.38

GOTREF

0.36

Bronnen

0.36

 TableColumn

0.35

IVEREF

0.35

 Frederick

0.35

Activations Density 0.002%

names of individuals

first names

No Comments

No Known Activations