INDEX

Explanations

terms related to context and inquiry

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

context in

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

:][

-0.44

MLLoader

-0.42

zaak

-0.41

withIdentifier

-0.40

ETHING

-0.37

ilistic

-0.37

trooper

-0.37

:`

-0.36

<!--

-0.36

tiken

-0.36

POSITIVE LOGITS

 CanadaChoose

0.52

BagConstraints

0.45

casila

0.39

 Throws

0.38

 كومونز

0.38

pushFollow

0.38

GrantedAuthority

0.37

 étoit

0.37

 мәкал

0.37

onedDateTime

0.36

Activations Density 0.001%

terms related to context and inquiry

context in

No Comments

No Known Activations