INDEX

Explanations

references to layout structures in a code context

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

programming/code tokens—especially mixed-case identifiers, ALL_CAPS constants, and API/library names within source or config snippets.

oai_token-act-pair · gpt-5 Triggered by @arielgoldsteinlab

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 kasarigan

-0.73

 kohdetta

-0.68

 BoxFit

-0.67

SharedCtor

-0.65

onViewCreated

-0.63

 Roskov

-0.60

دانشنامهٔ

-0.60

astéroïdes

-0.59

Tazama

-0.59

 <<<<<<<<<<<<<<

-0.58

POSITIVE LOGITS

 betweenstory

0.45

 jsPsych

0.42

➌

0.40

 selamat

0.39

 تانيه

0.37

yarnpkg

0.36

wurf

0.36

Reserved

0.35

 Administrativna

0.35

IActionResult

0.35

Activations Density 0.467%

references to layout structures in a code context

programming/code tokens—especially mixed-case identifiers, ALL_CAPS constants, and API/library names within source or config snippets.

No Comments

No Known Activations