INDEX

Explanations

patterns in structural or formatting elements of text

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

exercise psycho physician

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

 الحره

-0.75

findpost

-0.63

bootstrapcdn

-0.58

tagHelperRunner

-0.51

httphttps

-0.50

GEBURTS

-0.50

 Paglinawan

-0.48

 كومونز

-0.47

principalColumn

-0.46

 Meksiku

-0.46

POSITIVE LOGITS

▾

0.45

 Oliva

0.44

APR

0.43

<th>

0.42

 Gottlieb

0.42

Pengertian

0.42

 twilight

0.41

 touch

0.41

 Visser

0.41

rep

0.41

Activations Density 0.069%

patterns in structural or formatting elements of text

exercise psycho physician

No Comments

No Known Activations