INDEX

Explanations

copyright information and related metadata

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

academic paper metadata

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

UnitTesting

-0.56

Diweddarwch

-0.52

prepareStatement

-0.52

tagext

-0.51

WithMany

-0.51

 ब्रेकडाउन

-0.51

}],

-0.46

 kasarigan

-0.44

Portail

-0.44

:✨

-0.43

POSITIVE LOGITS

脚注の使い方

0.47

CppCodeGen

0.38

Slf

0.36

曖昧さ回避

0.34

setVerticalGroup

0.34

 елның

0.32

$_['

0.32

 мәгълү

0.31

 hulp

0.31

 GenerationType

0.31

Activations Density 0.155%

copyright information and related metadata

academic paper metadata

No Comments

No Known Activations