INDEX

Explanations

specific references to Roman history, organic chemistry, or cellular biology terms

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

Roman festivals

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

 كومونز

-0.44

AnchorTagHelper

-0.40

zarchiwizowane

-0.38

AddTagHelper

-0.38

 Schönheit

-0.37

addCriterion

-0.36

 dezelve

-0.35

 autorytatywna

-0.35

WriteTagHelper

-0.35

 importância

-0.35

POSITIVE LOGITS

 дописавши

0.51

 itſelf

0.51

__*/

0.50

 Exacts

0.49

 ſtand

0.48

 Roskov

0.48

ništvo

0.48

 leſs

0.47

Попис

0.46

 uſed

0.45

Activations Density 0.135%

specific references to Roman history, organic chemistry, or cellular biology terms

Roman festivals

No Comments

No Known Activations