INDEX

Explanations

occurrences of auxiliary verbs as well as closely related words

oai_token-act-pair · gemini-2.0-flash

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.22.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.22.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

 Underlying

-0.68

 termica

-0.63

 alluminio

-0.63

 fermés

-0.63

 picioare

-0.59

 engraçadas

-0.59

 colorés

-0.59

 étan

-0.59

 barbati

-0.58

 depoz

-0.57

POSITIVE LOGITS

 also

1.27

 considered

1.25

not

1.24

 used

1.20

 made

1.20

 taken

1.19

 seen

1.16

 shown

1.13

 held

1.13

 always

1.13

Activations Density 4.540%

occurrences of auxiliary verbs as well as closely related words

No Comments

No Known Activations