INDEX

Explanations

variations of the modal verb "can."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

PerformLayout

-0.75

 azzurro

-0.75

 layui

-0.70

 postres

-0.68

tieth

-0.68

GEBURTSDATUM

-0.68

Lähteet

-0.67

pence

-0.67

 userSchema

-0.66

 Affiliations

-0.66

POSITIVE LOGITS

can

1.60

CAN

1.40

Can

1.34

ocan

1.19

Can

1.18

can

1.16

be

1.09

 could

1.06

 easily

1.03

kan

0.96

Activations Density 0.240%

variations of the modal verb "can."

No Comments

No Known Activations