INDEX

Explanations

phrases indicating collaboration and research contributions within scientific and academic contexts

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

researchers, doctors, and institutes

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

NameInMap

-0.40

ことが

-0.38

事が

-0.36

doctype

-0.35

 Vordergrund

-0.34

îné

-0.34

 Vielen

-0.34

onReady

-0.33

 cielos

-0.33

Personensuche

-0.33

POSITIVE LOGITS

 decided

0.49

EndContext

0.48

出版年

0.48

rungsseite

0.47

 NSCoder

0.46

decided

0.46

 getConnection

0.45

 inwest

0.44

 inventors

0.43

tilhe

0.43

Activations Density 0.060%

phrases indicating collaboration and research contributions within scientific and academic contexts

researchers, doctors, and institutes

No Comments

No Known Activations