INDEX

Explanations

numerical patterns or sequences related to data and configuration information

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

oa̍t

-0.65

+#+#

-0.57

 незавершена

-0.54

 pełni

-0.52

ISupport

-0.52

TestBed

-0.51

 TType

-0.49

 nonUne

-0.49

 cetera

-0.48

enterOuterAlt

-0.48

POSITIVE LOGITS

day

0.69

day

0.68

Day

0.64

Day

0.63

 getClass

0.61

PhysRevD

0.60

DAY

0.60

 MessageType

0.57

days

0.57

DAY

0.56

Activations Density 0.453%

numerical patterns or sequences related to data and configuration information

No Comments

No Known Activations