INDEX

Explanations

phrases that express agreement or affirmation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 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

ніципалі

-0.51

 conmigo

-0.51

tagHelperRunner

-0.50

 plutôt

-0.49

setVerticalGroup

-0.48

 InputDecoration

-0.48

oredCriteria

-0.47

@"/

-0.46

 EconPapers

-0.46

⟭

-0.45

POSITIVE LOGITS

 Naturally

0.53

 naturally

0.49

Naturally

0.47

naturally

0.46

tru

0.44

yes

0.43

Yes

0.42

YES

0.40

そりゃ

0.40

Yes

0.40

Activations Density 0.272%

phrases that express agreement or affirmation

No Comments

No Known Activations