INDEX

Explanations

attends to programming-related tokens from related file or class 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

0:0.08

1:0.09

2:0.10

3:0.11

4:0.06

5:0.05

6:0.30

7:0.18

Negative Logits

Geplaatst

-0.46

ագրություններ

-0.45

脚注の使い方

-0.42

 protoimpl

-0.36

 ویکی‌پدیا

-0.35

DataMember

-0.35

كويكب

-0.34

सन्दर्भ

-0.34

AddHtmlAttribute

-0.34

PMailer

-0.33

POSITIVE LOGITS

illère

0.36

 resourceCulture

0.30

 Scra

0.24

̈́

0.23

󠁴

0.22

ี่ยม

0.21

Aut

0.21

Tô

0.21

=>'

0.21

Tó

0.21

Activations Density 0.002%

attends to programming-related tokens from related file or class tokens

No Comments

No Known Activations

attends to programming-related tokens from related file or class tokens

No Comments

No Known Activations