INDEX

Explanations

terms and phrases related to laws, penalties, and regulatory frameworks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

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

Negative Logits

 relâche

-0.41

cillor

-0.39

ٍ

-0.38

gettext

-0.38

őre

-0.35

 noDo

-0.35

 مفص

-0.35

 nakalista

-0.35

Jeudi

-0.33

elif

-0.32

POSITIVE LOGITS

ConstraintMaker

0.73

 mijne

0.64

LookAnd

0.60

xase

0.59

المناصب

0.54

yntaxException

0.52

 ddelweddau

0.52

 ActivatedRoute

0.52

:✨

0.52

 createSlice

0.51

Activations Density 1.930%

terms and phrases related to laws, penalties, and regulatory frameworks

No Comments

No Known Activations