INDEX

Explanations

elements related to data structures in programming

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

Code related to vertex normals

np_acts-logits-general · gemini-2.0-flash

vertex attributes and normals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 cortinas

-0.50

Prag

-0.47

 légale

-0.47

Begriffsklä

-0.47

asgi

-0.46

mergeFrom

-0.43

llas

-0.42

 ramas

-0.41

sdp

-0.40

dì

-0.40

POSITIVE LOGITS

 <<<<<<<<<<<<<<

0.68

 nakalista

0.64

 Roskov

0.64

MigrationBuilder

0.64

+#+

0.62

richTextPanel

0.60

 indices

0.60

 outsourcing

0.60

 face

0.58

 Gabb

0.58

Activations Density 0.120%

elements related to data structures in programming

Code related to vertex normals

vertex attributes and normals

No Comments

No Known Activations