INDEX

Explanations

references to places or contexts in a narrative

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 Lordships

-0.70

enumi

-0.62

OLUTION

-0.61

 MonoBehaviour

-0.60

feinander

-0.59

들에게

-0.59

ruimte

-0.58

OkHttpClient

-0.58

 sockaddr

-0.57

jahres

-0.57

POSITIVE LOGITS

 between

0.54

case

0.52

between

0.52

 nahilalakip

0.49

eed

0.49

 دى

0.48

 hindsight

0.47

 intégr

0.47

 Chwiliwch

0.45

__*/

0.44

Activations Density 0.360%

references to places or contexts in a narrative

No Comments

No Known Activations