INDEX

Explanations

references to positions, places, or roles within a hierarchy

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

abestanden

-0.91

ArrowToggle

-0.79

StoreMessageInfo

-0.78

 Италијани

-0.77

 مشين

-0.77

UVWXYZ

-0.77

providedIn

-0.75

twimg

-0.74

 متعلقه

-0.73

 kasarigan

-0.72

POSITIVE LOGITS

 place

3.14

place

2.64

 Place

2.44

Place

2.40

 PLACE

2.35

 places

2.28

PLACE

2.04

 lugar

2.01

places

1.96

 spot

1.90

Activations Density 0.229%

references to positions, places, or roles within a hierarchy

No Comments

No Known Activations