INDEX

Explanations

references to specific names or titles related to businesses, broadcasting networks, and publications

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

 GenerationType

-0.61

 Biôgrafia

-0.49

bootstrapcdn

-0.47

verwijspagina

-0.41

RenderAtEndOf

-0.41

 femininas

-0.40

 sluit

-0.40

 Lleva

-0.40

 fisuras

-0.40

Jîn

-0.40

POSITIVE LOGITS

참고

0.44

 iParam

0.43

lunda

0.40

 resourceCulture

0.39

最快更新

0.38

0.37

richTextPanel

0.37

 دریافت‌شده

0.37

Ecco

0.37

səhifə

0.36

Activations Density 0.166%

references to specific names or titles related to businesses, broadcasting networks, and publications

No Comments

No Known Activations