INDEX

Explanations

questions and requests for feedback in comments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

طلحات

-0.58

setVerticalGroup

-0.56

InvalidProtocol

-0.53

الحياه

-0.52

PREFERRED

-0.49

 toán

-0.48

transQ

-0.48

selaer

-0.48

,:,

-0.47

Ấ

-0.47

POSITIVE LOGITS

 comment

2.71

 comments

2.61

comment

2.37

 Comment

2.37

Comment

2.31

 Comments

2.22

comments

2.21

 commenting

2.21

 COMMENT

2.20

Comments

2.16

Activations Density 0.142%

questions and requests for feedback in comments

No Comments

No Known Activations