INDEX

Explanations

references to authors and their affiliations in academic contexts

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

authors of papers, abbreviations

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

seamnă

-0.50

 transfieras

-0.43

 noastre

-0.42

 Exactos

-0.41

enderror

-0.39

 posteriore

-0.38

 JpaRepository

-0.38

 trasparente

-0.38

 conclusiones

-0.37

__*/

-0.37

POSITIVE LOGITS

webElementXpaths

0.58

Portály

0.52

 utafitiHapana

0.48

NameInMap

0.47

verifyException

0.46

PSE

0.44

unk

0.44

graphical

0.44

Hochspringen

0.43

士

0.43

Activations Density 0.070%

references to authors and their affiliations in academic contexts

authors of papers, abbreviations

No Comments

No Known Activations