INDEX

Explanations

attends to tokens that relate to a web address or domain from tokens that reference a similar concept or topic

oai_attention-head · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-att/layer_0/width_16k/average_l0_61

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

0:0.03

1:0.01

2:0.01

3:0.02

4:0.02

5:0.01

6:0.02

7:0.01

8:0.70

9:0.02

10:0.01

11:0.01

12:0.02

13:0.02

14:0.01

15:0.01

Negative Logits

 zijne

-0.47

 ciudadanía

-0.45

 finanzas

-0.45

 desmotivaciones

-0.44

 Italij

-0.43

 civilización

-0.43

 suaminya

-0.43

 perfección

-0.42

 Infór

-0.42

 salvación

-0.42

POSITIVE LOGITS

APL

0.57

bib

0.56

CU

0.56

EK

0.54

Op

0.54

GEO

0.54

CT

0.53

BLE

0.52

BIB

0.52

Doc

0.52

Activations Density 0.020%

attends to tokens that relate to a web address or domain from tokens that reference a similar concept or topic

No Comments

No Known Activations