INDEX

Explanations

references to front-facing or anterior positions in various contexts

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

front or anterior parts

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 nahilalakip

-0.47

 southwestern

-0.47

 southwest

-0.46

kenstock

-0.46

 RemoteException

-0.45

 Administrativna

-0.44

 Farma

-0.44

 useParams

-0.43

antwoorden

-0.42

Southwest

-0.42

POSITIVE LOGITS

 front

0.82

front

0.77

Front

0.68

 Front

0.67

FRONT

0.64

 anteriore

0.63

 depan

0.61

 FRONT

0.59

前

0.56

 forward

0.54

Activations Density 0.032%

references to front-facing or anterior positions in various contexts

front or anterior parts

No Comments

No Known Activations