INDEX

Explanations

proper nouns, particularly names and surnames

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

ners

-0.61

 '\\;'

-0.54

findpost

-0.53

RegressionTest

-0.53

NERS

-0.52

Tikang

-0.51

Personendaten

-0.48

GHIJKLM

-0.47

//

-0.46

Nowhere

-0.45

POSITIVE LOGITS

訟

0.34

 transférez

0.34

struzioni

0.31

PlayerController

0.31

 TestBed

0.30

glEnable

0.30

pery

0.30

FragmentManager

0.30

blue

0.30

ga

0.30

Activations Density 0.292%

proper nouns, particularly names and surnames

No Comments

No Known Activations