INDEX

Explanations

phrases and constructs typically used in questions or inquiries

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

mathematical and multilingual instructions

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

脚注の使い方

-0.59

styleType

-0.58

StoryboardSegue

-0.55

twimg

-0.55

liosis

-0.54

出版年

-0.53

èdia

-0.53

andoah

-0.52

 AssemblyProduct

-0.51

Tē

-0.51

POSITIVE LOGITS

struzioni

0.40

◀

0.30

Seed

0.29

énéral

0.29

 bedien

0.29

 appelez

0.28

достатки

0.28

 prévention

0.28

曖昧さ回避

0.27

 seed

0.27

Activations Density 0.000%

phrases and constructs typically used in questions or inquiries

mathematical and multilingual instructions

No Comments

No Known Activations