INDEX

Explanations

references to legal or policy-related issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

ɵɵ

-0.47

Autoritní

-0.44

PerformLayout

-0.43

 betweenstory

-0.41

COUVER

-0.38

agi

-0.37

Depend

-0.37

typeDefinition

-0.35

 effic

-0.35

DEPEND

-0.34

POSITIVE LOGITS

ніципалі

0.63

 unrelated

0.60

 unconnected

0.50

 discul

0.49

 thematic

0.49

matically

0.48

 theme

0.48

 disconnected

0.47

Rüyada

0.47

 مشين

0.46

Activations Density 0.718%

references to legal or policy-related issues

No Comments

No Known Activations