INDEX

Explanations

questions and requests for information or clarification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 كومونز

-0.48

pholes

-0.46

mappedBy

-0.45

 onData

-0.40

 iNdEx

-0.40

 beginnetje

-0.38

 gevolg

-0.37

等你

-0.37

 Aggregate

-0.37

cipital

-0.36

POSITIVE LOGITS

 please

0.85

please

0.69

 PLEASE

0.65

 Please

0.60

Please

0.60

 пожалуйста

0.56

PLEASE

0.52

pls

0.52

plz

0.51

 Paglinawan

0.51

Activations Density 0.312%

questions and requests for information or clarification

No Comments

No Known Activations