INDEX

Explanations

references to the term "Big" along with its associated phrases or names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/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.36.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

談社

-0.41

hane

-0.39

دانشنامهٔ

-0.38

ကိုးကား

-0.37

tising

-0.36

Véxase

-0.36

 незавершена

-0.36

OVE

-0.35

DoubleQuotes

-0.34

merce

-0.34

POSITIVE LOGITS

big

1.02

big

0.95

BIG

0.93

BIG

0.93

Big

0.93

Big

0.86

wigs

0.70

 bang

0.69

gies

0.65

bang

0.64

Activations Density 0.083%

references to the term "Big" along with its associated phrases or names

No Comments

No Known Activations