INDEX

Explanations

terms associated with reputation or status in a social or culinary context

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

followed by prepositions or conjunctions

np_acts-logits-general · gemini-2.0-flash

abstract concepts and entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-1.21

-0.95

 which

-0.84

-0.73

",

-0.72

 vilket

-0.69

?,

-0.69

!,

-0.69

$,

-0.67

 hvilket

-0.67

POSITIVE LOGITS

<bos>

0.84

rungsseite

0.80

NUMX

0.79

NameInMap

0.76

 Paglinawan

0.74

 فريبيس

0.73

 $(\%)$

0.69

 AssemblyCulture

0.69

 ModelRenderer

0.67

OGND

0.67

Activations Density 0.819%

terms associated with reputation or status in a social or culinary context

followed by prepositions or conjunctions

abstract concepts and entities

No Comments

No Known Activations