INDEX

Explanations

interactions or conversations involving authority and subservience

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

Roman empire or Roman history

np_acts-logits-general · gemini-2.0-flash

Roman

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

PerformLayout

-0.66

 TextInputType

-0.64

 ویکی‌پدیا

-0.61

RegressionTest

-0.59

BeginInit

-0.56

 الرياضيه

-0.53

 الحره

-0.52

 Andante

-0.51

Skocz

-0.50

 Alamos

-0.49

POSITIVE LOGITS

 Roman

1.26

 Rome

1.10

Roman

1.09

 ROMAN

0.99

 Caesar

0.96

Rome

0.96

 Roma

0.96

 Romans

0.94

 roman

0.93

Rom

0.92

Activations Density 0.170%

interactions or conversations involving authority and subservience

Roman empire or Roman history

Roman

No Comments

No Known Activations