INDEX

Explanations

references to former officials and positions held by them

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

former roles or states

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

BagConstraints

-0.53

expandindo

-0.48

AxisAlignment

-0.44

 žád

-0.41

makeText

-0.41

 屋根

-0.40

 hırka

-0.39

styleType

-0.39

 MainAxisSize

-0.37

énario

-0.37

POSITIVE LOGITS

hoeddwyd

0.52

+#+

0.51

 former

0.49

Former

0.48

ioutil

0.47

 Former

0.46

EndGlobal

0.41

旧

0.41

 COMMENTS

0.41

archiviato

0.40

Activations Density 0.030%

references to former officials and positions held by them

former roles or states

No Comments

No Known Activations