INDEX

Explanations

references to specific locations and establishments within a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tagHelperRunner

-0.79

@"/

-0.56

isContained

-0.50

تقاوى

-0.50

новништво

-0.49

 Taktlose

-0.48

Jeografia

-0.48

Personendaten

-0.48

zeß

-0.47

fromnode

-0.47

POSITIVE LOGITS

 LCCN

0.47

 ſur

0.46

 ſta

0.43

 strdup

0.43

resizingMask

0.42

 faſt

0.42

KeySpec

0.41

SuppressLint

0.41

QFont

0.40

 ſtre

0.40

Activations Density 0.176%

references to specific locations and establishments within a narrative context

No Comments

No Known Activations