INDEX

Explanations

attends to the token "storage" from programming-related 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.12

2:0.13

3:0.15

4:0.11

5:0.10

6:0.13

7:0.14

Negative Logits

rrggbb

-0.26

 Pene

-0.26

ggars

-0.26

#}

-0.25

ształ

-0.25

stantial

-0.25

__.__

-0.25

jstor

-0.25

 ErrInvalid

-0.25

*/;

-0.25

POSITIVE LOGITS

multer

0.34

homonymie

0.33

хьтан

0.30

曖昧さ回避

0.29

artement

0.28

 BoxFit

0.28

identiality

0.27

 NgModule

0.27

tyfik

0.27

yarnpkg

0.27

Activations Density 0.000%

attends to the token "storage" from programming-related tokens

No Comments

No Known Activations