INDEX

Explanations

phrases indicating conditions and relationships between subjects

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

recommendations and obligations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 Administrativna

-0.63

astify

-0.60

OGND

-0.54

Tikang

-0.54

bcryptjs

-0.53

clerview

-0.51

 dAtA

-0.48

szönöm

-0.47

 ModelExpression

-0.47

 שוליים

-0.47

POSITIVE LOGITS

 recommends

0.46

べき

0.43

 sollten

0.42

tonode

0.42

hould

0.41

 should

0.41

 ought

0.40

apprendre

0.39

すべき

0.39

ควร

0.38

Activations Density 0.133%

phrases indicating conditions and relationships between subjects

recommendations and obligations

No Comments

No Known Activations