INDEX

Explanations

references to databases and their components

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

 cork

-0.69

 corks

-0.68

hyg

-0.68

 parlar

-0.67

Imit

-0.67

ctile

-0.67

 Pelleg

-0.65

τ

-0.63

shift

-0.63

 AppComponent

-0.62

POSITIVE LOGITS

 database

2.15

 Database

2.00

 databases

2.00

Database

1.78

 Databases

1.76

database

1.74

databases

1.66

Databases

1.59

 DATABASE

1.55

DATABASE

1.54

Activations Density 0.034%

references to databases and their components

No Comments

No Known Activations