INDEX

Explanations

references to coding or software documentation

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 19-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_19/width_16k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 الحره

-0.62

servable

-0.58

цездатний

-0.57

 שוליים

-0.57

 tartalomajánló

-0.52

 palla

-0.52

auro

-0.52

 betweenstory

-0.50

Viited

-0.50

 Paramount

-0.50

POSITIVE LOGITS

 noqa

0.59

 ModelExpression

0.57

 doInBackground

0.54

OCUS

0.49

toBeDefined

0.48

UCT

0.48

 الاطلاع

0.45

ISMS

0.44

туаль

0.43

Joy

0.42

Activations Density 0.175%

references to coding or software documentation

No Known Activations