INDEX

Explanations

questions and expressions of uncertainty

oai_token-act-pair · gpt-4o-mini Triggered by @bot

asking "what"

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

InjectAttribute

-0.74

JpaRepository

-0.54

 autorytatywna

-0.52

rlpool

-0.51

 schoenen

-0.49

featureID

-0.48

下载附件

-0.48

存于互联网档案馆

-0.48

farwyddwr

-0.48

 समीक्षाओं

-0.48

POSITIVE LOGITS

 CreateTagHelper

0.41

Ră

0.38

 disambiguazione

0.37

 resourceCulture

0.36

 emergency

0.35

Sab

0.35

ResourceType

0.34

ReturnType

0.33

 purpose

0.32

 Bohn

0.32

Activations Density 0.035%

questions and expressions of uncertainty

asking "what"

No Comments

No Known Activations