INDEX

Explanations

attends to specific action or description tokens from their associated contextual 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

0:0.45

1:0.18

2:0.13

3:0.04

4:0.03

5:0.02

6:0.05

7:0.06

Negative Logits

thâu

-0.48

 otomatig

-0.42

enderror

-0.42

 Taktlose

-0.41

็จ

-0.40

UnsafeEnabled

-0.39

 سكانية

-0.39

 متعلقه

-0.37

rungsseite

-0.36

 utafitiHapana

-0.35

POSITIVE LOGITS

getColumnIndex

0.39

PerformLayout

0.32

useNewUrlParser

0.31

strix

0.31

Allocated

0.31

 Subsequently

0.30

pportun

0.29

 Furthermore

0.29

 GoogleFonts

0.29

 Ronnie

0.28

Activations Density 0.406%

attends to specific action or description tokens from their associated contextual tokens

No Comments

No Known Activations