INDEX

Explanations

references to meals and dining experiences

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

meals

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

mybatisplus

-0.54

urlpatterns

-0.53

 protoimpl

-0.52

Rhestr

-0.51

tonode

-0.50

IUrlHelper

-0.50

wpi

-0.50

vânt

-0.49

 mouseY

-0.49

transQ

-0.48

POSITIVE LOGITS

 dinner

0.42

 lunch

0.40

 Lunch

0.39

 Dinner

0.38

 luncheon

0.37

Dinner

0.37

Lunch

0.35

 breakfast

0.33

 Unterkunft

0.32

 almuerzo

0.32

Activations Density 0.008%

references to meals and dining experiences

meals

No Comments

No Known Activations