INDEX

Explanations

high activation values associated with the beginning or introduction of content

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

segments of script or code, particularly focusing on syntax elements and commands.

oai_token-act-pair · gpt-4o Triggered by @timna

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.13.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

MLLoader

-0.68

Xna

-0.65

//</

-0.64

 مشين

-0.62

ագրություններ

-0.60

initializeApp

-0.60

Enllaços

-0.58

InjectAttribute

-0.58

LEncoder

-0.57

 clearColor

-0.57

POSITIVE LOGITS

 rubio

0.53

parole

0.50

 vader

0.49

 poesía

0.48

 opérateur

0.48

chengladbach

0.47

PSC

0.47

monix

0.46

压

0.46

...

0.46

Activations Density 0.425%

high activation values associated with the beginning or introduction of content

segments of script or code, particularly focusing on syntax elements and commands.

No Comments

No Known Activations