INDEX

Explanations

programming constructs related to thread management and function definitions

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

function call arguments

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

__(/*!

-0.59

saraba

-0.50

錬

-0.47

RTDA

-0.45

oinette

-0.45

annica

-0.44

delwed

-0.44

ioutil

-0.44

 خر

-0.43

Gdx

-0.43

POSITIVE LOGITS

ViewFeatures

0.43

ldots

0.41

 resourceCulture

0.41

Autoritní

0.40

 désolés

0.38

 contextLoads

0.37

+:+

0.36

 HasFactory

0.36

AsUp

0.36

 censiti

0.36

Activations Density 0.001%

programming constructs related to thread management and function definitions

function call arguments

No Comments

No Known Activations