INDEX

Explanations

phrases that include requests or polite directives

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

please followed by an action

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

OGND

-0.36

Achievements

-0.35

 existence

-0.35

 Crack

-0.34

awtextra

-0.33

 drugs

-0.33

DataContract

-0.32

berra

-0.32

 cracks

-0.31

 useSelector

-0.31

POSITIVE LOGITS

 please

0.67

rsiniz

0.65

 Bitte

0.61

IsMutable

0.60

Please

0.60

 Please

0.59

AddTagHelper

0.59

WriteTagHelper

0.57

 veuillez

0.56

Bitte

0.56

Activations Density 0.006%

phrases that include requests or polite directives

please followed by an action

No Comments

No Known Activations