INDEX

Explanations

phrases that express uncertainty or opinions

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

 fieldNum

-0.40

 organizza

-0.40

 coment

-0.36

 Typed

-0.35

łam

-0.35

區

-0.35

 Busse

-0.33

uted

-0.33

aupun

-0.32

endlich

-0.32

POSITIVE LOGITS

Autoritní

0.68

 Normdatei

0.59

Derbyniad

0.59

AndEndTag

0.59

CloseOperation

0.58

 SafeMath

0.56

 snippetHide

0.55

BufferException

0.55

TestingModule

0.54

ंदीखरीदारी

0.54

Activations Density 0.024%

phrases that express uncertainty or opinions

No Comments

No Known Activations