INDEX

Explanations

symbols and metrics related to data representation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

Personendaten

-1.29

Personensuche

-1.07

webElementXpaths

-0.97

 tartalomajánló

-0.88

تقاوى

-0.86

SharedDtor

-0.85

 kaarangay

-0.84

saraba

-0.80

 ffilmiau

-0.80

 disambiguazione

-0.79

POSITIVE LOGITS

 daarna

0.51

reszcie

0.47

وردار

0.46

כשיו

0.45

KEYCODE

0.45

Wasn

0.44

存于互联网档案馆

0.44

følgelig

0.43

 vervolgens

0.43

 dann

0.41

Activations Density 0.277%

symbols and metrics related to data representation

No Comments

No Known Activations