INDEX

Explanations

programming-related terms and functions

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-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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RenderAtEndOf

-0.79

extAlignment

-0.73

 betweenstory

-0.72

 '\\;'

-0.72

RegressionTest

-0.69

 CanadaChoose

-0.68

ंदीखरीदारी

-0.68

tvguidetime

-0.66

hyrchwyd

-0.65

GEBURTSDATUM

-0.65

POSITIVE LOGITS

 Phaser

0.73

Phaser

0.51

 lace

0.31

 monomer

0.31

en

0.30

 config

0.29

aqu

0.28

 Zahn

0.28

 psych

0.27

 rast

0.27

Activations Density 0.003%

programming-related terms and functions

No Comments

No Known Activations