INDEX

Explanations

attends to tokens indicating types or aspects from tokens that identify or categorize them

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

0:0.11

1:0.11

2:0.10

3:0.07

4:0.09

5:0.02

6:0.13

7:0.32

Negative Logits

 مشين

-0.41

ArgsConstructor

-0.40

GEBURTSDATUM

-0.39

Datuak

-0.39

astify

-0.36

TagHelper

-0.34

 numberWith

-0.32

 ujednoznacz

-0.32

UserScript

-0.32

 estekak

-0.31

POSITIVE LOGITS

('.');

0.35

 jsPsych

0.34

 "));

0.34

 '');

0.32

}},\

0.31

pumpe

0.31

'))

0.31

PerSecond

0.30

'',

0.30

ılmıştır

0.30

Activations Density 0.566%

attends to tokens indicating types or aspects from tokens that identify or categorize them

No Comments

No Known Activations