INDEX

Explanations

programming symbols and operators, particularly in a coding context

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

values and declarations

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

Negative Logits

 nahilalakip

-0.65

richTextPanel

-0.57

 propOrder

-0.54

GEBURTSDATUM

-0.53

Autoritní

-0.51

Personendaten

-0.50

omock

-0.48

TRAILING

-0.48

Lähteet

-0.48

 وتسجيلات

-0.47

POSITIVE LOGITS

 Lokales

0.47

 jälkeen

0.38

another

0.34

Vidite

0.33

 kaos

0.33

everywhere

0.32

 otra

0.32

こちらは

0.32

‍♂️

0.32

bland

0.31

Activations Density 0.083%

programming symbols and operators, particularly in a coding context

values and declarations

No Known Activations