INDEX

Explanations

patterns or sequences that resemble data or encoded characters

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

Tokens after currency codes and abbreviations

np_acts-logits-general · gemini-2.0-flash

signal identifiers

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

bro

-0.50

oogle

-0.48

 Shore

-0.47

STAN

-0.47

Constantes

-0.46

rister

-0.46

 them

-0.44

bro

-0.44

asi

-0.43

 free

-0.43

POSITIVE LOGITS

Personendaten

1.02

uxxxx

0.71

fjspx

0.71

UnusedPrivate

0.70

 jadx

0.64

 occuper

0.60

Datuak

0.60

orghini

0.58

 saurait

0.56

 ArrayAdapter

0.55

Activations Density 0.099%

patterns or sequences that resemble data or encoded characters

Tokens after currency codes and abbreviations

signal identifiers

No Comments

No Known Activations