INDEX

Explanations

hashtags and keywords related to specific topics or subjects

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

 Reſ

-0.64

 Anſ

-0.64

 faſt

-0.63

 Houſe

-0.62

 ſta

-0.62

 anſ

-0.62

 Inſ

-0.59

 juſt

-0.59

 reaſon

-0.58

 ſtate

-0.58

POSITIVE LOGITS

0.52

0.45

TabIndex

0.42

Disclaimer

0.40

Mil

0.37

NamedQueries

0.37

Mil

0.37

Me

0.36

amp

0.35

forall

0.35

Activations Density 0.120%

hashtags and keywords related to specific topics or subjects

No Comments

No Known Activations