INDEX

Explanations

references to specific methods or variables in structured data and programming

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

IntoConstraints

-1.26

ſelves

-1.22

ſelf

-1.20

 myſelf

-1.18

 pleaſure

-1.17

 purpoſe

-1.16

IsContent

-1.16

ſy

-1.14

 Monfieur

-1.12

 deſt

-1.11

POSITIVE LOGITS

0.89

0.79

and

0.75

0.74

is

0.70

of

0.69

or

0.67

in

0.66

on

0.65

0.62

Activations Density 0.812%

references to specific methods or variables in structured data and programming

No Comments

No Known Activations