INDEX

Explanations

technical terms and phrases related to electrical components and their functionalities

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

form noun-adjective compounds

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

сь

-0.38

icot

-0.34

gemeins

-0.34

kund

-0.33

ake

-0.32

发表于

-0.31

chi

-0.30

chapp

-0.30

enterOuterAlt

-0.29

blat

-0.29

POSITIVE LOGITS

 BoxFit

0.66

principalTable

0.66

########.

0.65

EDEFAULT

0.63

MessageOf

0.63

 Majefty

0.58

transQ

0.55

+#+

0.55

CppCodeGen

0.55

 lenker

0.55

Activations Density 0.061%

technical terms and phrases related to electrical components and their functionalities

form noun-adjective compounds

No Comments

No Known Activations