INDEX

Explanations

attends to the token "print" from a variety of contexts involving programming or output-related 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

0:0.78

1:0.01

2:0.01

3:0.02

4:0.08

5:0.02

6:0.03

7:0.02

Negative Logits

 <=",

-0.41

XMLSchema

-0.39

SequentialGroup

-0.38

awtextra

-0.38

ropractic

-0.35

GEBURTS

-0.35

 informée

-0.34

InSection

-0.32

 armband

-0.32

 Espagne

-0.32

POSITIVE LOGITS

ズル

0.30

:+:

0.28

みてください

0.27

])[

0.27

anager

0.27

lgari

0.27

Portale

0.26

 wireType

0.24

subscribed

0.24

Hauptartikel

0.24

Activations Density 0.015%

attends to the token "print" from a variety of contexts involving programming or output-related tokens

No Comments

No Known Activations