INDEX

Explanations

attends to self-referential tokens from tokens indicating relinquishment or limitation

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

0:0.70

1:0.01

2:0.01

3:0.02

4:0.15

5:0.03

6:0.01

7:0.03

Negative Logits

of

-0.36

 they

-0.32

adaptiveStyles

-0.31

tagHelperRunner

-0.31

she

-0.30

 while

-0.30

in

-0.30

исленность

-0.30

his

-0.29

POSITIVE LOGITS

ſelves

0.50

 houſe

0.48

 itſelf

0.48

 ſmall

0.48

 myſelf

0.46

 pleaſure

0.46

 greateſt

0.44

 Monfieur

0.44

 ſtate

0.43

 ourselves

0.43

Activations Density 0.005%

attends to self-referential tokens from tokens indicating relinquishment or limitation

No Comments

No Known Activations