INDEX

Explanations

elements, variables, and other notations used in mathematical expressions and coding

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

Mathematical or scientific text

np_acts-logits-general · gemini-2.0-flash

chemical structures and equations

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

),

-0.29

enc

-0.28

 Reef

-0.28

 hers

-0.27

Reference

-0.27

 hoá

-0.27

foe

-0.26

 iArr

-0.26

 adher

-0.26

Ref

-0.26

POSITIVE LOGITS

 transférez

0.80

TypedDataSet

0.72

 mergeFrom

0.66

 transfieras

0.63

StoryboardSegue

0.57

 CanadaChoose

0.57

 kasarigan

0.56

TestingModule

0.56

AndEndTag

0.56

 utafitiHapana

0.55

Activations Density 1.611%

elements, variables, and other notations used in mathematical expressions and coding

Mathematical or scientific text

chemical structures and equations

No Comments

No Known Activations