INDEX

Explanations

attends to tokens that include specified professions or notable identities from linked tokens that represent activities or contexts

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

0:0.09

1:0.10

2:0.07

3:0.12

4:0.09

5:0.06

6:0.24

7:0.19

Negative Logits

存于互联网档案馆

-0.52

 Савезне

-0.41

 MainAxisSize

-0.40

abetes

-0.40

κτηρισ

-0.40

DoubleQuotes

-0.40

 صوتيه

-0.39

 المعيارى

-0.39

 initComponents

-0.39

errHandler

-0.39

POSITIVE LOGITS

"]));

0.29

ffilm

0.29

 Hawkins

0.27

 Gehir

0.26

 kjø

0.25

 EnglishChoose

0.25

 Dwyer

0.24

faßt

0.24

zeugen

0.24

>[]

0.23

Activations Density 0.216%

attends to tokens that include specified professions or notable identities from linked tokens that represent activities or contexts

No Comments

No Known Activations