INDEX

Explanations

names of people and places, especially those from the Middle East

oai_token-act-pair · gemini-2.0-flash

say "p" words

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

verwijspagina

-0.43

cault

-0.38

localctx

-0.38

nalités

-0.35

ChromeDriver

-0.35

CROSSTALK

-0.34

iniums

-0.34

сылки

-0.34

departamento

-0.33

oredCriteria

-0.33

POSITIVE LOGITS

0.85

per

0.74

0.71

ps

0.67

pt

0.67

par

0.66

pat

0.66

bu

0.65

bs

0.64

pas

0.64

Activations Density 12.669%

names of people and places, especially those from the Middle East

say "p" words

No Comments

No Known Activations

names of people and places, especially those from the Middle East

say "p" words

No Comments

No Known Activations