INDEX

Explanations

words and phrases related to computation and programming contexts

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

an antipsychotic

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

 ProtoMessage

-0.42

VYMaps

-0.42

LookAnd

-0.41

 sientes

-0.41

yntaxException

-0.41

 zatr

-0.40

 isolado

-0.39

 lccn

-0.39

 penjual

-0.38

ConstraintMaker

-0.38

POSITIVE LOGITS

 Administrativna

0.54

dur

0.50

vide

0.48

Portail

0.47

+#+#

0.47

hton

0.47

 Armin

0.47

otor

0.46

forestry

0.46

senic

0.46

Activations Density 0.019%

words and phrases related to computation and programming contexts

an antipsychotic

No Comments

No Known Activations