INDEX

Explanations

structured categories and classifications within datasets

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

introduces a list

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 fédé

-0.45

Espèce

-0.40

 noDo

-0.40

 répon

-0.40

parsedMessage

-0.39

]--;

-0.38

 otomatig

-0.36

},[])

-0.35

 térmico

-0.35

 estético

-0.35

POSITIVE LOGITS

 estekak

0.52

 Normdatei

0.50

клопе

0.50

including

0.50

 including

0.49

 yaitu

0.48

 Tsub

0.47

 např

0.46

特别是

0.46

 například

0.46

Activations Density 0.070%

structured categories and classifications within datasets

introduces a list

No Comments

No Known Activations