INDEX

Explanations

technical terms and specific applications related to devices and systems

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

sentence endings

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

tvguidetime

-0.63

 Signalez

-0.59

+#+#

-0.58

#+#

-0.56

évaluateur

-0.55

rungsseite

-0.54

 legit

-0.51

httphttps

-0.50

transQ

-0.47

 autorytatywna

-0.47

POSITIVE LOGITS

:✨

0.79

tanleria

0.69

 económico

0.44

Aholisi

0.42

 ISTAT

0.42

 éter

0.41

 einverstanden

0.40

GEBURTSDATUM

0.40

ఔ

0.40

WireFormatLite

0.39

Activations Density 1.560%

technical terms and specific applications related to devices and systems

sentence endings

No Comments

No Known Activations