INDEX

Explanations

personal pronouns and references to individuals

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

Follows interrogative words

np_acts-logits-general · gemini-2.0-flash

did not, is not

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

RenderAtEndOf

-0.56

 AssemblyCulture

-0.55

Personensuche

-0.54

 Administrativna

-0.52

AddTagHelper

-0.52

⤹

-0.50

ConstraintMaker

-0.50

MethodManager

-0.49

 Italijani

-0.48

 autorytatywna

-0.47

POSITIVE LOGITS

styleable

0.39

帖最后由

0.36

Thank

0.35

 Sicht

0.34

 Wieder

0.34

 vocês

0.34

So

0.33

Remember

0.32

 अंत

0.32

<h4>

0.32

Activations Density 0.628%

personal pronouns and references to individuals

Follows interrogative words

did not, is not

No Comments

No Known Activations