INDEX

Explanations

expressions related to personal recommendations and support

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 enforced

-0.48

 Schmid

-0.47

 Stamm

-0.47

 balles

-0.46

 Schmitz

-0.46

 Tiefen

-0.44

 plumas

-0.43

blick

-0.43

 Metzger

-0.43

 risa

-0.43

POSITIVE LOGITS

oredCriteria

0.79

 мәкал

0.77

Хьажоргаш

0.77

出版年

0.74

Personensuche

0.73

uxxxx

0.73

OGND

0.73

 Roskov

0.73

хьтан

0.72

httphttps

0.71

Activations Density 0.006%

expressions related to personal recommendations and support

No Comments

No Known Activations