INDEX

Explanations

programming-related syntax and structure

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

Code snippets' initializations or configurations

np_acts-logits-general · gemini-2.0-flash

code identifiers like lin_

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

Vidite

-0.85

省市镇

-0.66

 snippetHide

-0.65

цездатний

-0.63

rrggbb

-0.62

WriteTagHelper

-0.62

MigrationBuilder

-0.62

 {}));

-0.60

Jeografia

-0.60

kháu

-0.58

POSITIVE LOGITS

 виправивши

0.52

 Enzymes

0.52

 Akismet

0.50

izr

0.50

 Massa

0.49

 initState

0.49

 Kicks

0.47

 Perfume

0.47

bookmarks

0.47

OPHY

0.47

Activations Density 0.033%

programming-related syntax and structure

Code snippets' initializations or configurations

code identifiers like lin_

No Comments

No Known Activations