INDEX

Explanations

references to plans and planning terminology

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

 مشين

-0.94

EDEFAULT

-0.91

NameInMap

-0.90

yarnpkg

-0.80

DeleteBehavior

-0.79

__':

-0.79

 esternos

-0.77

XmlAccessorType

-0.72

IsContent

-0.71

//-->

-0.70

POSITIVE LOGITS

 plans

0.80

 Plans

0.74

 PLAN

0.68

Plans

0.66

 PLANS

0.66

plans

0.64

PLAN

0.58

 ahead

0.57

kton

0.53

 planners

0.51

Activations Density 0.212%

references to plans and planning terminology

No Comments

No Known Activations