INDEX

Explanations

proper nouns, especially names of people and specific players

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 Sorry

-0.53

 Claire

-0.52

 Lucy

-0.51

genheit

-0.51

ietal

-0.50

 Betsy

-0.49

icylic

-0.48

 Kate

-0.48

🏻

-0.48

iligten

-0.47

POSITIVE LOGITS

MENAFN

0.74

NewUrlParser

0.65

Andre

0.63

 Terrell

0.61

 Biôgrafia

0.61

 Marquis

0.60

Jereo

0.60

0.58

رشف

0.58

 NUKAT

0.58

Activations Density 0.263%

proper nouns, especially names of people and specific players

No Comments

No Known Activations