INDEX

Explanations

instances of formatting or markup elements in the text

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

input from "user" label

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

 Grüsse

-0.46

principalTable

-0.41

 raiſ

-0.39

 Nachbarn

-0.39

 ſte

-0.39

gebäude

-0.38

IsPostBack

-0.37

sstelle

-0.36

 deſt

-0.35

 raising

-0.35

POSITIVE LOGITS

:✨

0.88

 ویکی‌پدی

0.57

OGND

0.54

 Italijanski

0.49

mergeFrom

0.47

IndentedString

0.47

 Meksiku

0.47

хьтан

0.46

Rohy

0.46

 Autorisations

0.46

Activations Density 0.000%

instances of formatting or markup elements in the text

input from "user" label

No Comments

No Known Activations