INDEX

Explanations

phrases related to support, empowerment, and action-oriented terms

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

application and use

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

 disambiguazione

-0.80

Hentet

-0.79

 Chwiliwch

-0.73

queryInterface

-0.72

 ddelweddau

-0.69

 doInBackground

-0.69

 فريبيس

-0.69

GOTREF

-0.68

parsedMessage

-0.67

 CreateTagHelper

-0.67

POSITIVE LOGITS

着

0.37

 Index

0.35

跳

0.35

 thus

0.34

一行

0.34

Index

0.33

处

0.33

 nakalista

0.32

getIndex

0.32

code

0.31

Activations Density 0.084%

phrases related to support, empowerment, and action-oriented terms

application and use

No Comments

No Known Activations