INDEX

Explanations

code-related chunks of text that involve licenses and software terms

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

LLocation

-0.58

はじめに

-0.56

saraba

-0.53

Билгалдахарш

-0.48

лтемелер

-0.47

WithIOException

-0.47

isContained

-0.46

ersch

-0.46

자

-0.46

المناصب

-0.46

POSITIVE LOGITS

TabIndex

0.88

 AssemblyCompany

0.57

SizeMode

0.55

WEBPACK

0.55

 ویکی‌پدیای

0.53

jsii

0.52

jspx

0.50

 verlassen

0.50

olol

0.49

phosa

0.49

Activations Density 0.039%

code-related chunks of text that involve licenses and software terms

No Comments

No Known Activations