INDEX

Explanations

attends to the token denoted "1" from the context provided in earlier tokens

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

0:0.10

1:0.10

2:0.05

3:0.07

4:0.23

5:0.21

6:0.09

7:0.12

Negative Logits

esgue

-0.32

までも

-0.32

phosa

-0.30

 naman

-0.30

ıntı

-0.30

 theirs

-0.30

↵↵

-0.30

 calloc

-0.29

 renon

-0.29

 männis

-0.28

POSITIVE LOGITS

 виправивши

0.60

 متعلقه

0.55

 فريبيس

0.54

styleType

0.54

InSection

0.49

 AssemblyCulture

0.48

 <=",

0.48

 المعيارى

0.48

@"/

0.47

PerformLayout

0.46

Activations Density 0.023%

attends to the token denoted "1" from the context provided in earlier tokens

No Comments

No Known Activations