INDEX

Explanations

references to individuals who are retired or formerly active in various professions

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

retired, former, abandoned, fallen

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

 arran

-0.36

pró

-0.35

 giebt

-0.35

timents

-0.34

 grew

-0.33

 introduce

-0.32

next

-0.32

 gethan

-0.31

 starts

-0.31

REAM

-0.30

POSITIVE LOGITS

SharedDtor

0.69

 retired

0.63

 former

0.59

internalType

0.58

 veter

0.58

Retired

0.57

 Retired

0.57

 Former

0.56

 للمعارف

0.55

 يتيمه

0.54

Activations Density 0.024%

references to individuals who are retired or formerly active in various professions

retired, former, abandoned, fallen

No Comments

No Known Activations