INDEX

Explanations

references to Washington, DC

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

TypedDataSet

-0.48

InputTagHelper

-0.42

}{@

-0.42

CodedInputStream

-0.39

 attachment

-0.38

('-'

-0.38

:^(

-0.37

 seriously

-0.37

modb

-0.37

BagConstraints

-0.36

POSITIVE LOGITS

DC

0.63

DC

0.60

att

0.58

dc

0.50

 Neurosci

0.48

dc

0.48

 artykuł

0.47

Controllo

0.46

 saint

0.45

 Eust

0.44

Activations Density 0.056%

references to Washington, DC

No Comments

No Known Activations