INDEX

Explanations

attends to various skills-related tokens from associated context 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

0:0.74

1:0.01

2:0.02

3:0.02

4:0.13

5:0.01

6:0.01

7:0.03

Negative Logits

İstinadlar

-0.27

tonsoft

-0.26

pti

-0.26

 Kasih

-0.25

gång

-0.24

кисло

-0.24

 Walkover

-0.24

 SPDX

-0.23

jalá

-0.23

liesslich

-0.23

POSITIVE LOGITS

StoryboardSegue

0.49

Hentet

0.43

verwijspagina

0.41

 defStyleAttr

0.40

+:+

0.39

makeConstraints

0.38

AxisAlignment

0.38

expandindo

0.37

 AssemblyVersion

0.36

epik

0.36

Activations Density 0.008%

attends to various skills-related tokens from associated context tokens

No Comments

No Known Activations