INDEX

Explanations

attends to quotation marks around phrases from later tokens that reference those phrases

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

0:0.10

1:0.09

2:0.14

3:0.04

4:0.04

5:0.02

6:0.08

7:0.44

Negative Logits

"?>

-0.33

zane

-0.26

 tkinter

-0.26

OBO

-0.26

 nakalista

-0.26

 forstå

-0.26

Gim

-0.26

Hod

-0.25

 Pard

-0.25

ơn

-0.25

POSITIVE LOGITS

RegressionTest

0.33

ContentAsync

0.32

SerializedSize

0.30

EndContext

0.29

 porcelaine

0.29

 насељу

0.28

 acrylique

0.28

 cref

0.28

quate

0.28

 quæ

0.28

Activations Density 0.596%

attends to quotation marks around phrases from later tokens that reference those phrases

No Comments

No Known Activations