INDEX

Explanations

terms related to control and regulation

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

getYear

-0.51

 February

-0.45

 YEAR

-0.45

 pierres

-0.45

January

-0.45

February

-0.44

 ficam

-0.44

 Fernanda

-0.44

 January

-0.42

 février

-0.42

POSITIVE LOGITS

control

1.34

 control

1.32

 Control

1.26

 CONTROL

1.15

Control

1.13

ontrol

1.04

CONTROL

1.00

 Controls

0.99

 controls

0.97

 controll

0.97

Activations Density 0.113%

terms related to control and regulation

No Known Activations