INDEX

Explanations

function definitions in programming code

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

code definitions and method names

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

 ennemi

-0.36

 mijne

-0.36

 profesión

-0.36

 poème

-0.33

 geschiedenis

-0.32

dialement

-0.32

 encuesta

-0.31

 vieill

-0.31

 péri

-0.31

 oorsp

-0.30

POSITIVE LOGITS

the

0.63

 Pears

0.61

 retrieving

0.61

ushes

0.61

 Koop

0.60

rona

0.60

 retrieve

0.60

ajs

0.60

wiliwch

0.59

 shrinks

0.57

Activations Density 0.002%

function definitions in programming code

code definitions and method names

No Comments

No Known Activations