INDEX

Explanations

references to the second-person perspective and the pronoun "you."

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

where followed by a pronoun

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

 lendemain

-0.32

NSNotification

-0.31

ientôt

-0.29

TIMEOUT

-0.28

Còn

-0.27

 Kedua

-0.27

 Neub

-0.27

 redenen

-0.26

the

-0.26

GORITH

-0.24

POSITIVE LOGITS

Dónde

0.85

OGND

0.84

 where

0.84

 InputDecoration

0.81

 Where

0.80

 где

0.79

Where

0.79

where

0.78

 donde

0.77

 όπου

0.77

Activations Density 0.037%

references to the second-person perspective and the pronoun "you."

where followed by a pronoun

No Comments

No Known Activations