INDEX

Explanations

questions and statements related to uncertainty or contemplation

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

specific terms or foreign words

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

-0.54

ligkeit

-0.47

sbericht

-0.44

Décès

-0.44

infall

-0.44

 workshop

-0.43

îner

-0.43

INE

-0.42

finch

-0.41

 Workshop

-0.41

POSITIVE LOGITS

saraba

0.54

évaluateur

0.48

 Theſe

0.43

Източници

0.42

 ordinaires

0.39

లాలు

0.39

 genoeg

0.38

 كومونز

0.38

 whoſe

0.37

 enough

0.36

Activations Density 0.071%

questions and statements related to uncertainty or contemplation

specific terms or foreign words

No Comments

No Known Activations