INDEX

Explanations

repetitions of the word "many."

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

VIAF

-0.42

 Crusoe

-0.40

spunk

-0.39

ITC

-0.39

 vườn

-0.39

 waistcoat

-0.38

spro

-0.38

isation

-0.38

 Orbital

-0.38

brate

-0.37

POSITIVE LOGITS

 many

1.25

 MANY

1.16

 Many

1.09

MANY

1.08

many

1.04

Many

1.02

ToMany

0.90

Banyak

0.87

 muchos

0.85

Beaucoup

0.85

Activations Density 0.049%

repetitions of the word "many."

No Comments

No Known Activations