INDEX

Explanations

references to reports or essays

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

"a/an" followed by a noun

np_acts-logits-general · gemini-2.0-flash

in a statement

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

ViewFeatures

-1.04

aarrggbb

-1.02

الإنجليزية

-0.86

 متعلقه

-0.86

expandindo

-0.86

ItemBackground

-0.86

 Meksiku

-0.84

 estekak

-0.82

 createState

-0.80

 otomatig

-0.79

POSITIVE LOGITS

 recent

1.07

 letter

0.94

 article

0.91

recent

0.79

 récente

0.73

 email

0.73

 reciente

0.69

 tweet

0.67

 report

0.66

 blog

0.64

Activations Density 0.434%

references to reports or essays

"a/an" followed by a noun

in a statement

No Comments

No Known Activations