INDEX

Explanations

attends to descriptions or attributes indicated with double asterisks from related segments or sections that are indicated with double square brackets

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

0:0.10

1:0.19

2:0.27

3:0.08

4:0.11

5:0.07

6:0.05

7:0.09

Negative Logits

hoeddwyd

-0.30

 Rossetti

-0.27

MigrationBuilder

-0.25

styleable

-0.24

né

-0.24

 encephalitis

-0.24

 CreateTagHelper

-0.24

Δείτε

-0.24

 Cowley

-0.23

 argc

-0.23

POSITIVE LOGITS

rawDesc

0.39

berdayakan

0.38

 quelcon

0.35

 جغرافيا

0.35

مصادر

0.34

DotNetBar

0.33

المناصب

0.31

FormTagHelper

0.30

 Hamlin

0.30

antaranya

0.30

Activations Density 0.046%

attends to descriptions or attributes indicated with double asterisks from related segments or sections that are indicated with double square brackets

No Comments

No Known Activations