INDEX

Explanations

attends to the word "also" from various tokens that provide additional information or context

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

0:0.36

1:0.18

2:0.16

3:0.05

4:0.02

5:0.01

6:0.03

7:0.14

Negative Logits

unknownFields

-0.28

StructEnd

-0.28

ARY

-0.26

 Lightboxes

-0.26

 onAnimation

-0.26

 noDo

-0.25

 CURIAM

-0.25

AutoresizingMask

-0.25

Gemeinsame

-0.24

lably

-0.24

POSITIVE LOGITS

\{\\

0.23

 simplifié

0.23

 оригіналу

0.22

brainly

0.22

คร

0.21

bruck

0.21

طفى

0.21

liefer

0.20

 educativas

0.19

 ancla

0.19

Activations Density 0.360%

attends to the word "also" from various tokens that provide additional information or context

No Comments

No Known Activations