INDEX

Explanations

references to specific people, names, or titles in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 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

 فيها

-0.46

 فيه

-0.45

 Paglinawan

-0.44

 אנו

-0.44

onSearch

-0.43

rašymas

-0.40

 ModelExpression

-0.40

<?=$

-0.39

而去

-0.39

isenberg

-0.39

POSITIVE LOGITS

bootstrapcdn

0.82

Geplaatst

0.75

quiao

0.74

ResumeLayout

0.73

Vidite

0.70

 transfieras

0.69

terase

0.68

#+#

0.68

TintMode

0.68

ChildScrollView

0.67

Activations Density 1.538%

references to specific people, names, or titles in various contexts

No Comments

No Known Activations