INDEX

Explanations

coding constructs related to variable assignment and object creation

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

document.createElement

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

TypedDataSet

-0.69

 kaarangay

-0.56

keyColumn

-0.56

uxxxx

-0.52

exitRule

-0.50

jsii

-0.49

ArgsConstructor

-0.48

RTEE

-0.48

❷

-0.48

 propOrder

-0.48

POSITIVE LOGITS

RefNanny

0.42

 oorspron

0.39

sproz

0.38

 tärke

0.38

 äldre

0.38

 ProtoMessage

0.37

 łą

0.37

COMMAND

0.36

esterno

0.36

 laaj

0.36

Activations Density 0.012%

coding constructs related to variable assignment and object creation

document.createElement

No Comments

No Known Activations