INDEX

Explanations

punctuation and formatting cues related to data presentation

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

text following numbers

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

OGND

-0.52

RTSC

-0.51

Gön

-0.51

 aapt

-0.44

Chham

-0.42

raisals

-0.40

DrawerToggle

-0.40

Diweddarwch

-0.39

apter

-0.39

alse

-0.39

POSITIVE LOGITS

SequentialGroup

0.48

 Wikiseite

0.46

Sziasztok

0.43

 imprimée

0.40

➌

0.39

ioterapia

0.38

 fjor

0.38

Notae

0.38

 biały

0.38

Activations Density 0.025%

punctuation and formatting cues related to data presentation

text following numbers

No Comments

No Known Activations