INDEX

Explanations

the word "you" in various contexts

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

you think or know

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

WebElementEntity

-0.33

 verbund

-0.33

due

-0.33

 Partagez

-0.32

동

-0.32

pyplot

-0.32

 jednost

-0.31

UN

-0.31

-0.30

POSITIVE LOGITS

الحياه

0.60

 Autorisations

0.58

ьажоргаш

0.56

 propOrder

0.56

niſſe

0.56

 beſch

0.54

 témoig

0.54

 dieſem

0.54

 imagui

0.54

 geſch

0.53

Activations Density 0.043%

the word "you" in various contexts

you think or know

No Comments

No Known Activations