INDEX

Explanations

elements related to programming constructs and mathematical notation

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

macro definitions and mathematical notation

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

 betweenstory

-0.83

orsese

-0.56

adaptiveStyles

-0.56

lighet

-0.51

 lard

-0.50

ագրություններ

-0.50

 vire

-0.50

 popcorn

-0.49

 Realtor

-0.49

閖

-0.49

POSITIVE LOGITS

цездатний

0.52

mergeFrom

0.52

WriteTagHelper

0.41

 ModelExpression

0.38

isContained

0.36

UrlResolution

0.36

Hentet

0.35

tagHelper

0.34

 getItemId

0.34

 videre

0.32

Activations Density 0.200%

elements related to programming constructs and mathematical notation

macro definitions and mathematical notation

No Comments

No Known Activations