INDEX

Explanations

attends to specific method calls or programming terms from related terms that suggest a context or functionality

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

0:0.33

1:0.18

2:0.15

3:0.08

4:0.06

5:0.03

6:0.04

7:0.07

Negative Logits

basicConfig

-0.30

IUrlHelper

-0.25

ίν

-0.25

ỡng

-0.24

IContainer

-0.24

parsedMessage

-0.24

Pab

-0.23

 step

-0.23

Jad

-0.23

 stessi

-0.22

POSITIVE LOGITS

AnimationsModule

0.23

estival

0.22

 typelib

0.21

뀜

0.20

 autorytatywna

0.20

fromCharCode

0.20

 poignée

0.20

mybatisplus

0.20

 Osiris

0.20

pyplot

0.19

Activations Density 0.541%

attends to specific method calls or programming terms from related terms that suggest a context or functionality

No Comments

No Known Activations