INDEX

Explanations

expressions of preference or recommendation

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

Non-English words or code snippets

np_acts-logits-general · gemini-2.0-flash

certainly recommend or gained

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

iprot

-0.47

•

-0.45

The

-0.43

apad

-0.43

bboxes

-0.42

}).

-0.42

 Then

-0.41

ieniu

-0.41

ädie

-0.41

insp

-0.41

POSITIVE LOGITS

 nahilalakip

0.76

ReusableCell

0.71

 архивлан

0.67

Hochspringen

0.67

Personendaten

0.63

 Wikimedijinoj

0.62

 futurs

0.61

 صوتيه

0.61

 future

0.60

قایناقلار

0.60

Activations Density 0.049%

expressions of preference or recommendation

Non-English words or code snippets

certainly recommend or gained

No Comments

No Known Activations