INDEX

Explanations

the word "again" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

kano

-0.85

 Hale

-0.71

 Swain

-0.67

 Purdy

-0.66

ZI

-0.66

tisseur

-0.65

 fers

-0.64

Koc

-0.63

matsu

-0.62

што

-0.62

POSITIVE LOGITS

Again

1.36

 Again

1.33

 again

1.33

 AGAIN

1.31

again

1.30

AGAIN

1.26

 igjen

1.03

jsdelivr

1.03

 Lagi

0.99

 Audiodateien

0.93

Activations Density 0.033%

the word "again" in various contexts

No Comments

No Known Activations