INDEX

Explanations

references to data structures or formats in text

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

academic paper links

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

 المعيارى

-0.53

antd

-0.51

NOPQRST

-0.50

othed

-0.42

]=>

-0.40

➯

-0.40

Spoljašnje

-0.39

aarrggbb

-0.39

zw

-0.39

 <<<<<<<<<<<<<<

-0.39

POSITIVE LOGITS

 EconPapers

0.46

 ConstraintSet

0.44

 descarga

0.43

 représenter

0.42

føl

0.42

 undergraduates

0.42

ValueStyle

0.41

adaptiveStyles

0.41

 añadido

0.41

 estekak

0.40

Activations Density 0.004%

references to data structures or formats in text

academic paper links

No Comments

No Known Activations