INDEX

Explanations

locations and sites related to events or activities

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

locations and venues

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

OGND

-0.69

 '\\;'

-0.60

 Autorizaciones

-0.59

KommentareTeilen

-0.54

Notae

-0.51

Personendaten

-0.50

 EconPapers

-0.49

 invokingState

-0.49

exitRule

-0.48

 Wallflower

-0.48

POSITIVE LOGITS

 venue

0.42

 lägen

0.42

 venues

0.41

 locations

0.39

 location

0.39

 flyg

0.38

 apartments

0.38

 annex

0.38

 gewä

0.37

位於

0.37

Activations Density 0.017%

locations and sites related to events or activities

locations and venues

No Comments

No Known Activations