INDEX

Explanations

proper nouns, particularly names and titles of individuals in various contexts

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

names and titles

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

 fær

-0.36

 stories

-0.35

 verhalen

-0.35

 optique

-0.33

 óptica

-0.32

ⓧ

-0.32

 novios

-0.32

 пожалуйста

-0.32

ographies

-0.32

 svých

-0.31

POSITIVE LOGITS

 nakalista

0.72

RefNanny

0.66

 NSCoder

0.57

ViewImports

0.57

PerformLayout

0.57

BeginInit

0.55

脚注の使い方

0.52

mbggenerated

0.50

InjectAttribute

0.49

posedge

0.48

Activations Density 0.011%

proper nouns, particularly names and titles of individuals in various contexts

names and titles

No Comments

No Known Activations