INDEX

Explanations

keywords related to measurement and evaluation

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

ending in "-er" that describe people or things

np_acts-logits-general · gemini-2.0-flash

descriptor and agent nouns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 peindre

-0.96

 défendre

-0.87

OrBuilder

-0.79

 فريبيس

-0.77

 remercier

-0.75

 voyons

-0.74

 boire

-0.71

 décrire

-0.71

 courir

-0.70

SharedCtor

-0.70

POSITIVE LOGITS

 confirmer

0.61

 mover

0.53

 eater

0.53

prefixer

0.52

 changer

0.51

ferrer

0.51

IFIER

0.51

uploader

0.51

 remover

0.50

 studier

0.50

Activations Density 0.834%

keywords related to measurement and evaluation

ending in "-er" that describe people or things

descriptor and agent nouns

No Comments

No Known Activations