INDEX

Explanations

attends to tokens related to specific keywords, such as "p," "browser," "test," and "email," from more general function 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.42

1:0.14

2:0.09

3:0.06

4:0.08

5:0.02

6:0.06

7:0.10

Negative Logits

IsContent

-0.30

例文帳に追加

-0.27

})}\

-0.26

CodeAttribute

-0.26

%%%%%%%%%%

-0.26

)))),

-0.26

});

-0.25

 Kome

-0.25

She

-0.25

vanni

-0.25

POSITIVE LOGITS

 Walkover

0.33

Население

0.32

зулта

0.30

ArrowToggle

0.30

 nahilalakip

0.28

 Gehen

0.28

MLLoader

0.28

 חיצוניים

0.27

ValueGeneration

0.27

 Dider

0.27

Activations Density 0.957%

attends to tokens related to specific keywords, such as "p," "browser," "test," and "email," from more general function tokens

No Comments

No Known Activations