INDEX

Explanations

attends to tokens that represent outcomes or states from preceding tokens that signify conditions 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

0:0.21

1:0.24

2:0.18

3:0.05

4:0.04

5:0.02

6:0.05

7:0.16

Negative Logits

Brainz

-0.38

 Lokales

-0.37

NAL

-0.35

 snippetHide

-0.35

 المعيارى

-0.32

□

-0.32

Filmo

-0.30

uxxxx

-0.30

 חיצוניים

-0.30

POSITIVE LOGITS

mybatisplus

0.30

Lähteet

0.28

ագրություններ

0.27

তথ্যসূত্র

0.27

 terakhir

0.26

IGraphics

0.26

ourg

0.26

Referències

0.25

by

0.25

 Hill

0.24

Activations Density 0.466%

attends to tokens that represent outcomes or states from preceding tokens that signify conditions or contexts

No Comments

No Known Activations