INDEX

Explanations

key actions and events in a sports context

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

preceding numbers or scores

np_acts-logits-general · gemini-2.0-flash

performance outcomes and rankings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 mußten

-0.65

 např

-0.56

umy

-0.55

olev

-0.53

 läßt

-0.52

囉

-0.51

unif

-0.51

 klein

-0.51

fag

-0.50

 müßte

-0.49

POSITIVE LOGITS

 stuttering

0.64

ècie

0.63

 belated

0.62

KommentareTeilen

0.58

 Hopf

0.58

 saites

0.57

Hochspringen

0.57

DeleteBehavior

0.56

himovic

0.56

 autorytatywna

0.56

Activations Density 0.138%

key actions and events in a sports context

preceding numbers or scores

performance outcomes and rankings

No Comments

No Known Activations