INDEX

Explanations

references to subdivisions or subcategories within a broader context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

RTLD

-0.59

Билгалдахарш

-0.45

RTDA

-0.43

GOTREF

-0.41

 surla

-0.40

 demografica

-0.40

ztal

-0.38

ViewSet

-0.38

řad

-0.38

 Autorizaciones

-0.37

POSITIVE LOGITS

 subsidiary

0.66

 subgroup

0.66

 subcommittee

0.62

 subgroups

0.57

sub

0.57

 subset

0.56

\{\\

0.55

RenderAtEndOf

0.55

 subcategory

0.53

 subsidi

0.53

Activations Density 1.203%

references to subdivisions or subcategories within a broader context

No Comments

No Known Activations