INDEX

Explanations

terms related to photons and their interactions in various experimental setups

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

photons and particles

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

hand

-0.35

meld

-0.35

 csal

-0.33

ful

-0.33

 раскры

-0.32

ﾉ

-0.32

 apellidos

-0.32

 muka

-0.31

Cif

-0.31

 mata

-0.30

POSITIVE LOGITS

 disambiguazione

0.76

 AssemblyCulture

0.74

 المعيارى

0.60

adpleegd

0.59

contentLoaded

0.55

 الحره

0.55

parsedMessage

0.54

IVEREF

0.53

UrlResolution

0.53

المناصب

0.52

Activations Density 0.121%

terms related to photons and their interactions in various experimental setups

photons and particles

No Comments

No Known Activations