INDEX

Explanations

names of geographical features and landmarks

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

terase

-0.39

ganu

-0.38

 jsPsych

-0.38

Opéra

-0.37

 éxitos

-0.33

detectChanges

-0.32

autorest

-0.31

 unimaginable

-0.31

Faça

-0.31

inige

-0.31

POSITIVE LOGITS

rungsseite

0.69

oneofs

0.60

 InputDecoration

0.58

 surla

0.58

 ModelExpression

0.52

 Italijani

0.51

vyšší

0.51

sizeCache

0.51

principalTable

0.50

 peak

0.50

Activations Density 0.033%

names of geographical features and landmarks

No Comments

No Known Activations