INDEX

Explanations

interrogative phrases indicating a request for information or guidance

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

Follows a question

np_acts-logits-general · gemini-2.0-flash

common answer starters

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

optionalTypeArgs

-0.41

 avoient

-0.38

 demikian

-0.38

sû

-0.36

 arşivlendi

-0.36

 réjou

-0.36

 pouvoit

-0.35

 mérit

-0.35

Kaynakça

-0.35

 khas

-0.35

POSITIVE LOGITS

 autorytatywna

0.82

\{\\

0.69

 يتيمه

0.56

">—

0.56

 ***!

0.53

ResumeLayout

0.52

下载附件

0.52

Biografie

0.51

 "..\..\

0.51

IPAC

0.49

Activations Density 0.635%

interrogative phrases indicating a request for information or guidance

Follows a question

common answer starters

No Comments

No Known Activations