INDEX

Explanations

attends to the second token from various self-referential tokens

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

0:0.10

1:0.11

2:0.11

3:0.07

4:0.05

5:0.02

6:0.12

7:0.38

Negative Logits

enumii

-0.28

WriteAttribute

-0.28

ininger

-0.26

 discorso

-0.24

 bata

-0.23

 HttpNotFound

-0.23

 pesky

-0.23

Elev

-0.23

rella

-0.23

 Koop

-0.23

POSITIVE LOGITS

XmlAccessType

0.36

 kasarigan

0.36

 betweenstory

0.36

FormTagHelper

0.35

Portail

0.35

 дописавши

0.34

endpush

0.33

gdx

0.33

__':

0.33

WindowConstants

0.32

Activations Density 0.416%

attends to the second token from various self-referential tokens

No Comments

No Known Activations