INDEX

Explanations

underscore-prefixed identifiers, likely related to programming or variable naming conventions

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

user input

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

 Inscrivez

-0.50

ContentAlignment

-0.48

tagext

-0.45

 tertulis

-0.44

msgSender

-0.43

ględ

-0.43

 rallado

-0.42

 moeite

-0.41

antaranya

-0.41

 CreateTagHelper

-0.41

POSITIVE LOGITS

Rhestr

0.66

@"/

0.52

KommentareTeilen

0.50

richTextPanel

0.50

spo

0.48

Nub

0.48

bigr

0.48

roben

0.48

Ser

0.47

 Cloud

0.47

Activations Density 0.038%

underscore-prefixed identifiers, likely related to programming or variable naming conventions

user input

No Comments

No Known Activations