INDEX

Explanations

occurrences of placeholders or delimiters in a structured format

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

followed by specific words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

saw

-0.63

 HttpContext

-0.62

перь

-0.62

Etsy

-0.61

riwal

-0.59

 MenuView

-0.58

migrationBuilder

-0.58

ện

-0.57

Linki

-0.57

:%.

-0.57

POSITIVE LOGITS

IsMutable

0.78

lagi

0.61

BindView

0.58

wack

0.57

 &___

0.57

 tufted

0.56

 Grüsse

0.56

createComponent

0.56

ufficient

0.55

cellation

0.54

Activations Density 0.008%

occurrences of placeholders or delimiters in a structured format

followed by specific words

No Comments

No Known Activations