INDEX

Explanations

definitions and instances of the word "define" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Money

-0.50

passenger

-0.49

money

-0.48

Passenger

-0.46

 ater

-0.45

milliseconds

-0.45

passengers

-0.44

Passengers

-0.43

Mouse

-0.42

trom

-0.42

POSITIVE LOGITS

 Define

1.13

 Defines

1.05

 definition

1.03

 Definition

1.01

 defines

0.99

Define

0.99

 define

0.98

 defining

0.97

 defin

0.96

 definitions

0.96

Activations Density 0.151%

definitions and instances of the word "define" in various contexts

No Comments

No Known Activations