INDEX

Explanations

the beginning of a new text or section

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

Follows "this", comma, or hashtag

np_acts-logits-general · gemini-2.0-flash

recipe

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

المكان

-0.58

DrawerToggle

-0.51

Cordialement

-0.50

 ernst

-0.49

 թվական

-0.49

chaff

-0.49

pens

-0.49

LLocation

-0.47

 plastiques

-0.47

Prensa

-0.46

POSITIVE LOGITS

 recipe

1.46

 recipes

1.25

recipe

1.22

 Recipe

1.09

Recipe

1.03

 Rezept

0.98

recipes

0.98

 recette

0.96

 RECIPE

0.96

 receta

0.93

Activations Density 0.153%

the beginning of a new text or section

Follows "this", comma, or hashtag

recipe

No Comments

No Known Activations