INDEX

Explanations

increment operations within code

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

 queſta

-0.98

 beſte

-0.88

 ſei

-0.88

ſſung

-0.86

ロウィン

-0.83

 ſeiner

-0.80

 ſua

-0.80

ésultats

-0.79

<unused14>

-0.79

<unused3>

-0.79

POSITIVE LOGITS

0.36

GOTREF

0.33

++;

0.31

UseVisualStyle

0.31

 however

0.28

..

0.27

EndContext

0.25

 Umum

0.25

0.24

분

0.24

Activations Density 0.003%

increment operations within code

No Comments

No Known Activations