INDEX

Explanations

phrases related to assessment and categorization of issues or claims

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 lecz

-0.60

 мәкал

-0.56

 transfieras

-0.56

 Италијани

-0.54

InjectMocks

-0.54

 attire

-0.53

optionalTypeArgs

-0.53

 nahilalakip

-0.53

исленность

-0.53

ACHUSET

-0.52

POSITIVE LOGITS

 weirdly

0.62

NameInMap

0.57

そういう

0.53

 cosas

0.52

 thing

0.50

 biß

0.48

 commonest

0.47

 allerlei

0.47

 stuff

0.47

 things

0.45

Activations Density 4.957%

phrases related to assessment and categorization of issues or claims

No Comments

No Known Activations