INDEX

Explanations

attends to educational tokens marked with ** from unspecified or generic tokens marked with [[ ]]

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-att-16k

Configuration

google/gemma-scope-2b-pt-att/layer_0/width_16k/average_l0_104

Prompts (Dashboard)

36,864 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

0:0.02

1:0.55

2:0.05

3:0.04

4:0.03

5:0.21

6:0.03

7:0.03

Negative Logits

 itſelf

-1.44

 myſelf

-1.43

Efq

-1.38

 ―――――

-1.27

 Monfieur

-1.24

 Theſe

-1.23

 Jefus

-1.19

ſelf

-1.18

^(@)

-1.17

 doubtnut

-1.17

POSITIVE LOGITS

0.75

↵↵

0.73

<eos>

0.68

0.66

or

0.65

in

0.61

and

0.59

on

0.57

to

0.57

Activations Density 0.049%

attends to educational tokens marked with ** from unspecified or generic tokens marked with [[ ]]

No Comments

No Known Activations