INDEX

Explanations

references and citations related to scientific articles and publications

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

publication citations

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

 bezeichneter

-0.46

Personendaten

-0.45

::_('

-0.44

 виправивши

-0.44

 critic

-0.42

bitat

-0.39

 Norden

-0.38

 воро

-0.38

 թվական

-0.38

qtype

-0.37

POSITIVE LOGITS

 Signalez

0.48

PyExc

0.47

 Chwiliwch

0.47

 وتسجيلات

0.46

BeginInit

0.45

 propOrder

0.44

 Autorizaciones

0.44

Hentet

0.44

basicConfig

0.43

WebVitals

0.43

Activations Density 0.007%

references and citations related to scientific articles and publications

publication citations

No Comments

No Known Activations