INDEX

Explanations

references to governmental authority and administrative actions

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

Titles or positions of people

np_acts-logits-general · gemini-2.0-flash

official titles and roles

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

{#

-0.64

 betweenstory

-0.55

IVEREF

-0.52

springframework

-0.51

 laiton

-0.50

 contribué

-0.50

ereco

-0.49

 Indians

-0.49

tonsoft

-0.49

 péri

-0.49

POSITIVE LOGITS

脚注の使い方

0.70

DoubleQuotes

0.67

 district

0.60

 District

0.58

erate

0.57

Collector

0.56

District

0.54

 Collector

0.53

 MonoBehaviour

0.53

 officer

0.52

Activations Density 0.059%

references to governmental authority and administrative actions

Titles or positions of people

official titles and roles

No Comments

No Known Activations