INDEX

Explanations

attends to action-related tokens from tokens indicating subsequent context or details

oai_attention-head · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-att-16k

Configuration

google/gemma-scope-2b-pt-att/layer_10/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.10.attn.hook_z

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.10

2:0.10

3:0.08

4:0.05

5:0.02

6:0.17

7:0.37

Negative Logits

Personensuche

-0.40

 Drapeau

-0.35

 تضيفلها

-0.33

WithIOException

-0.33

PRNewswire

-0.32

ítése

-0.31

 giudi

-0.30

άζ

-0.30

Zunanje

-0.30

 AssemblyCulture

-0.29

POSITIVE LOGITS

ceğini

0.23

frag

0.23

 fila

0.23

 Timoteo

0.23

 smelt

0.23

 OkHttpClient

0.23

nac

0.23

⮕

0.23

LabelTagHelper

0.23

Seb

0.23

Activations Density 0.145%

attends to action-related tokens from tokens indicating subsequent context or details

No Comments

No Known Activations