INDEX

Explanations

references to regulations or rules within a community context

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

Comes after words indicating possibility/desire/quality

np_acts-logits-general · gemini-2.0-flash

high throughput, high resolution

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

 nakalista

-0.65

errHandler

-0.65

出版年

-0.60

 ujednoznacz

-0.60

 surla

-0.60

ContentAlignment

-0.60

ViewFeatures

-0.59

rrggbb

-0.58

ValueStyle

-0.56

InitVars

-0.56

POSITIVE LOGITS

çe

0.32

RTCK

0.31

 paste

0.30

Portail

0.29

面上

0.29

sok

0.28

 bolsillo

0.27

 tribus

0.27

 pocket

0.27

 trick

0.27

Activations Density 1.665%

references to regulations or rules within a community context

Comes after words indicating possibility/desire/quality

high throughput, high resolution

No Comments

No Known Activations