INDEX

Explanations

instances of positional values in a data structure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

Efq

-1.12

 متعلقه

-1.06

 Monfieur

-1.03

 autorytatywna

-1.03

InjectAttribute

-1.01

 photolibrary

-1.00

 Theſe

-0.95

 ARXIV

-0.94

 doubtnut

-0.94

$_"

-0.94

POSITIVE LOGITS

pos

2.59

pos

2.49

Pos

2.42

Pos

2.30

POS

1.95

POS

1.88

getPos

1.39

 position

1.36

 positions

1.35

 posi

1.34

Activations Density 0.041%

instances of positional values in a data structure

No Comments

No Known Activations