INDEX

Explanations

proper nouns, specifically names of individuals or organizations

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

 surla

-0.69

ValueStyle

-0.67

 Италијани

-0.62

 שוליים

-0.58

 autorytatywna

-0.57

AccessorTable

-0.55

Життєпис

-0.54

 незавершена

-0.54

Tikang

-0.53

 AssemblyCompany

-0.52

POSITIVE LOGITS

 gyvento

0.41

 gloire

0.36

send

0.33

 kıs

0.31

 gydy

0.31

 decision

0.31

 amarillas

0.31

ActionPerformed

0.31

 solution

0.31

 fact

0.31

Activations Density 0.010%

proper nouns, specifically names of individuals or organizations

No Comments

No Known Activations