INDEX

Explanations

references to "control" in various contexts

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

control and associated terms

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

getYear

-0.56

 February

-0.49

 cartera

-0.47

 dumne

-0.47

February

-0.47

 ficam

-0.47

January

-0.47

 Meksi

-0.46

 février

-0.46

 novios

-0.45

POSITIVE LOGITS

 control

1.37

control

1.37

 Control

1.31

 CONTROL

1.20

Control

1.19

ontrol

1.07

 controle

1.05

CONTROL

1.02

 controll

1.00

 controls

0.95

Activations Density 0.046%

references to "control" in various contexts

control and associated terms

No Comments

No Known Activations