INDEX

Explanations

proper nouns and names of locations related to court cases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

enis

-0.47

 lhes

-0.45

zuk

-0.45

şen

-0.45

 olvido

-0.44

crate

-0.44

 evolução

-0.44

 industriel

-0.43

Citiți

-0.43

 Letter

-0.42

POSITIVE LOGITS

 Minnesota

0.93

 متعلقه

0.92

CppMethod

0.87

Minnesota

0.86

 Minn

0.83

Minneapolis

0.76

 Duluth

0.76

DebuggerNonUser

0.75

Tracce

0.74

 ligiloj

0.74

Activations Density 1.674%

proper nouns and names of locations related to court cases

No Comments

No Known Activations

proper nouns and names of locations related to court cases

No Comments

No Known Activations