INDEX

Explanations

phrases and references to authorship in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

 læng

-0.52

 toegang

-0.46

 dignité

-0.45

parsedMessage

-0.45

 péné

-0.44

 partij

-0.43

 rijden

-0.43

 igualdad

-0.43

 femenino

-0.42

 flèche

-0.42

POSITIVE LOGITS

帖最后由

0.62

0.55

Aiheesta

0.53

__(/*!

0.51

/**

0.51

0.50

`{.

0.47

 <=",

0.47

WEBPACK

0.47

 kasarigan

0.47

Activations Density 0.643%

phrases and references to authorship in various contexts

No Comments

No Known Activations