INDEX

Explanations

references to political figures and their actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

and

-0.46

te

-0.40

-0.39

 ordinaires

-0.37

 italiane

-0.37

for

-0.36

 refusé

-0.36

as

-0.35

POSITIVE LOGITS

rungsseite

1.25

findpost

1.12

 disambiguazione

1.03

ngdoc

1.01

AccessorTable

1.01

RepeatedField

1.00

 bezeichneter

0.98

WithIOException

0.96

Tikang

0.95

UnusedPrivate

0.94

Activations Density 0.352%

references to political figures and their actions

No Comments

No Known Activations