INDEX

Explanations

attends to formatting-related tokens from tokens that follow, specifically focusing on formatting syntax such as appending or displaying values

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

0:0.15

1:0.11

2:0.08

3:0.05

4:0.04

5:0.04

6:0.10

7:0.38

Negative Logits

:✨

-0.56

 Wikimedijinoj

-0.34

 beginnetje

-0.28

TagMode

-0.28

çait

-0.26

শে

-0.24

ɵ

-0.24

تقاوى

-0.24

KURZBESCHREIBUNG

-0.23

PLEMENT

-0.23

POSITIVE LOGITS

WithIOException

0.40

DoubleQuotes

0.33

 BoxFit

0.28

ormais

0.28

BoxLayout

0.27

uasaan

0.27

ogast

0.27

Grüsse

0.27

OPSIS

0.27

Activations Density 0.111%

attends to formatting-related tokens from tokens that follow, specifically focusing on formatting syntax such as appending or displaying values

No Comments

No Known Activations