INDEX

Explanations

phrases indicating comparisons or evaluations

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

InsertCommand

-0.39

onews

-0.39

 pedest

-0.35

ValueStyle

-0.34

law

-0.33

rimp

-0.33

นด์

-0.32

ценка

-0.32

fromnode

-0.32

ymal

-0.32

POSITIVE LOGITS

dafx

0.70

 సౌకర్య

0.53

новништво

0.49

 تضيفلها

0.49

 typelib

0.49

ViewImports

0.48

 insuffisamment

0.48

 surla

0.46

 الرياضيه

0.46

 Walkover

0.44

Activations Density 0.087%

phrases indicating comparisons or evaluations

No Comments

No Known Activations