INDEX

Explanations

terms related to explanation and clarity in communication

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

 مشين

-0.39

 joaat

-0.39

ignées

-0.35

 controversies

-0.34

Auteur

-0.33

balleur

-0.33

 dealerships

-0.32

 Roskov

-0.32

 bilhões

-0.32

 democrats

-0.31

POSITIVE LOGITS

 nahilalakip

0.57

➟

0.55

SequentialGroup

0.54

:✨

0.50

EndContext

0.50

 Geſ

0.50

 mijne

0.50

ImageContext

0.50

 intptr

0.48

 незавершена

0.47

Activations Density 0.092%

terms related to explanation and clarity in communication

No Comments

No Known Activations