INDEX

Explanations

references to formal classifications and listings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

aarrggbb

-0.54

 Normdatei

-0.52

 mxArray

-0.52

رشف

-0.45

 ويكيپيديا

-0.43

 للاسماء

-0.43

Hentet

-0.42

Citiți

-0.42

Geplaatst

-0.42

MemoryWarning

-0.41

POSITIVE LOGITS

Carriera

0.46

 ProtoMessage

0.42

 Diſ

0.41

 dinosau

0.41

ApiController

0.40

 durs

0.40

 SEDS

0.40

ніципалі

0.40

itens

0.40

 Conſ

0.40

Activations Density 0.005%

references to formal classifications and listings

No Comments

No Known Activations