INDEX

Explanations

statements or phrases indicating contradictions or discrepancies between claims and reality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

setVerticalGroup

-0.73

 estekak

-0.71

SourceChecksum

-0.67

 Administrativna

-0.65

AnchorStyles

-0.65

دانشنامهٔ

-0.63

majánló

-0.62

DockStyle

-0.62

 utafitiHapana

-0.61

<unused14>

-0.60

POSITIVE LOGITS

/*

0.38

stylers

0.32

im

0.31

Халык

0.27

zustellen

0.26

IS

0.25

ctivas

0.25

ższe

0.24

ッド

0.24

থ্য

0.24

Activations Density 1.234%

statements or phrases indicating contradictions or discrepancies between claims and reality

No Comments

No Known Activations