INDEX

Explanations

references to figures or illustrations in academic or technical documents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_65k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.17.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

曖昧さ回避

-0.60

AsUp

-0.54

 announced

-0.52

grav

-0.51

 JTextField

-0.51

 noted

-0.50

 gron

-0.50

MarshalTo

-0.49

#![

-0.48

TInt

-0.48

POSITIVE LOGITS

 تانيه

0.72

 doubtnut

0.71

|}{$

0.65

 للاسماء

0.65

出版年

0.63

 kantoor

0.60

 NDEBUG

0.60

✨:

0.60

 înc

0.59

 depositphotos

0.59

Activations Density 0.008%

references to figures or illustrations in academic or technical documents

No Comments

No Known Activations