INDEX

Explanations

references to figures, tables, or citations in a document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

GEBURTSDATUM

-0.61

rungsseite

-0.58

Gambas

-0.56

ELTS

-0.51

elemField

-0.51



-0.50

WireFormatLite

-0.50

styleType

-0.50

 linkovi

-0.50

 typelib

-0.48

POSITIVE LOGITS

oo

0.54

۰۰

0.52

OO

0.49

doo

0.45

ooo

0.44

OOO

0.44

OO

0.42

COO

0.42

оо

0.42

yelidikan

0.42

Activations Density 0.800%

references to figures, tables, or citations in a document

No Comments

No Known Activations