INDEX

Explanations

attends to concluding tokens that often follow introductory 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

0:0.11

1:0.17

2:0.13

3:0.17

4:0.14

5:0.04

6:0.09

7:0.11

Negative Logits

CodedInputStream

-0.42

 quæ

-0.29

GIA

-0.29

脚注の使い方

-0.28

 secours

-0.28

ásra

-0.28

Viitteet

-0.28

 intervenir

-0.28

gląda

-0.27

getAdapter

-0.27

POSITIVE LOGITS

AccessorTable

0.33

rungsseite

0.29

ruvate

0.27

Composable

0.27

 calendriers

0.27

ANTLR

0.26

 censiti

0.26

่านั้น

0.25

Hauptartikel

0.24

bitField

0.24

Activations Density 0.091%

attends to concluding tokens that often follow introductory tokens

No Comments

No Known Activations