INDEX

Explanations

instances of specific phrases or concepts related to progression and changes in states or conditions

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

Bib

-0.43

 CHtml

-0.43

lien

-0.39

Linden

-0.38

guin

-0.38

 High

-0.38

Matchers

-0.36

 Jump

-0.36

Bin

-0.36

 Forest

-0.36

POSITIVE LOGITS

 Chwiliwch

0.60

0.58

ंदीखरीदारी

0.53

GEBURTSDATUM

0.50

rouvez

0.47

fjspx

0.47

Jeografia

0.47

interopRequire

0.47

 esternos

0.46

 виправивши

0.45

Activations Density 0.001%

instances of specific phrases or concepts related to progression and changes in states or conditions

No Comments

No Known Activations