INDEX

Explanations

question marks throughout the text

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

bytes values

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

ValueStyle

-0.62

tagHelperRunner

-0.56

makeText

-0.52

 nahilalakip

-0.52

tagHelper

-0.52

providedIn

-0.50

 surla

-0.49

ToAction

-0.48

 useStyles

-0.47

WindowConstants

-0.47

POSITIVE LOGITS

MigrationBuilder

0.65

Personendaten

0.52

 cómod

0.46

 cómodo

0.45

 ouverts

0.44

izarse

0.44

silueta

0.43

irvana

0.42

 keper

0.42

 compréhen

0.42

Activations Density 0.076%

question marks throughout the text

bytes values

No Comments

No Known Activations