INDEX

Explanations

terms related to parity and classification of numbers

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 للمعارف

-0.84

 InputDecoration

-0.76

RegistryLite

-0.70

adomo

-0.65

Crumb

-0.62

 referenties

-0.60

SPJ

-0.60

 hierarchies

-0.58

 Beecher

-0.57

writeFieldEnd

-0.57

POSITIVE LOGITS

ẵn

1.02

 parity

0.60

AddTagHelper

0.60

ODD

0.59

 alternating

0.59

odd

0.59

UseVisualStyle

0.57

짝

0.57

odd

0.56

 ModelExpression

0.56

Activations Density 0.069%

terms related to parity and classification of numbers

No Comments

No Known Activations