INDEX

Explanations

modal verbs expressing ability or potential actions

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

PerformLayout

-0.90

 saites

-0.74

 layui

-0.72

pence

-0.71

WriteTagHelper

-0.69

Données

-0.69

fromnode

-0.68

libft

-0.68

tieth

-0.68

GEBURTSDATUM

-0.68

POSITIVE LOGITS

can

1.34

can

1.34

Can

1.34

CAN

1.34

Can

1.26

CAN

1.05

 easily

1.03

ocan

1.03

 Puede

1.02

 puede

1.00

Activations Density 0.819%

modal verbs expressing ability or potential actions

No Comments

No Known Activations