INDEX

Explanations

programming-related control flow statements and return values

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

RegressionTest

-0.74

Datuak

-0.69

featureID

-0.67

runApp

-0.64

SpringBootTest

-0.64

IVEREF

-0.64

-0.63

Personendaten

-0.62

AutoModerator

-0.61

 beginnetje

-0.61

POSITIVE LOGITS

 متعلقه

0.67

 utafitiHapana

0.65

FORME

0.56

 OMITTED

0.53

jonen

0.53

ReusableCell

0.51

θεν

0.50

setQuery

0.50

 snippetHide

0.49

arh

0.49

Activations Density 0.049%

programming-related control flow statements and return values

No Comments

No Known Activations