INDEX

Explanations

quantitative and mathematical expressions or terms

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

Hentet

-0.65

Бахар

-0.47

 ModelExpression

-0.47

 betweenstory

-0.47

pexpr

-0.46

adaptiveStyles

-0.46

matchCondition

-0.45

StoryboardSegue

-0.45

脚注の使い方

-0.44

 defaultstate

-0.44

POSITIVE LOGITS

<>());

0.48

verwijspagina

0.42

RUnlock

0.42

 navideño

0.41

 aislado

0.41

 navideña

0.40

 OMITBAD

0.40

บ้าง

0.40

有过

0.39

 testigos

0.39

Activations Density 2.236%

quantitative and mathematical expressions or terms

No Comments

No Known Activations