INDEX

Explanations

activities related to eating and meals

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

before or after meals

np_acts-logits-general · gemini-2.0-flash

eating meals

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 بيها

-0.62

 wireType

-0.52

 actionTypes

-0.52

kuuta

-0.48

 butterknife

-0.48

 الحره

-0.47

ніципа

-0.46

pinMode

-0.46

seits

-0.45

 Administrativna

-0.45

POSITIVE LOGITS

 lunch

2.12

 dinner

2.01

 breakfast

1.80

lunch

1.65

 Lunch

1.63

 supper

1.57

dinner

1.56

 Dinner

1.55

Lunch

1.53

Dinner

1.52

Activations Density 0.150%

activities related to eating and meals

before or after meals

eating meals

No Comments

No Known Activations