INDEX

Explanations

locations and geographical information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ویکی‌پدی

-0.52

новништво

-0.50

 Wrocław

-0.49

 raiſ

-0.49

$_['

-0.48

 psych

-0.47

 Balboa

-0.47

 kaarangay

-0.47

Referencie

-0.46

APOLIS

-0.46

POSITIVE LOGITS

 village

0.48

village

0.45

Village

0.42

 Village

0.41

 Settlement

0.41

 villages

0.39

settlement

0.38

Personensuche

0.38

Settlement

0.37

 villagers

0.36

Activations Density 0.178%

locations and geographical information

No Comments

No Known Activations