INDEX

Explanations

attends from tokens emphasizing specific topics or methods to tokens referring to statistical or causal inference concepts

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

0:0.35

1:0.14

2:0.06

3:0.05

4:0.05

5:0.04

6:0.11

7:0.15

Negative Logits

stateProvider

-0.36

WEBPACK

-0.36

 wikipagina

-0.35

ueses

-0.35

 saites

-0.35

modelBuilder

-0.35

 كومونز

-0.34

enumi

-0.32

 ExecuteAsync

-0.32

 azar

-0.32

POSITIVE LOGITS

principalTable

0.45

webElementXpaths

0.31

archiviato

0.30

ometric

0.30

Abbiamo

0.29

StoryboardSegue

0.29

يميديا

0.28

曖昧さ回避

0.28

homonymie

0.28

igshid

0.27

Activations Density 0.122%

attends from tokens emphasizing specific topics or methods to tokens referring to statistical or causal inference concepts

No Comments

No Known Activations