INDEX

Explanations

numerical values and their context in relation to races or competitions

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

numbers and measurements

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

évaluateur

-0.55

rungsseite

-0.54

Jegyzetek

-0.54

 createSlice

-0.51

ագրություններ

-0.51

 EconPapers

-0.49

Erreferentziak

-0.48

 AssemblyCompany

-0.48

aronder

-0.47

ñora

-0.47

POSITIVE LOGITS

 length

0.45

 الرياضيه

0.42

gam

0.42

mid

0.42

 Length

0.41

 midi

0.40

 shorter

0.38

 hilly

0.38

гата

0.37

 stadium

0.37

Activations Density 0.000%

numerical values and their context in relation to races or competitions

numbers and measurements

No Comments

No Known Activations