INDEX

Explanations

links and references, with a focus on variation and sustainability

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

Code, copyright, or technical jargon

np_acts-logits-general · gemini-2.0-flash

link to

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

他和

-0.47

chuhe

-0.46

ault

-0.45

 ftill

-0.45

 absentee

-0.41

 himſelf

-0.41

 offerts

-0.41

gulan

-0.41

 télé

-0.40

 Burrows

-0.40

POSITIVE LOGITS

contentLoaded

0.82

 những

0.78

 aquellos

0.73

########.

0.72

 resourceCulture

0.69

 aquellas

0.68

interopRequire

0.67

những

0.67

 selecting

0.67

 those

0.66

Activations Density 0.514%

links and references, with a focus on variation and sustainability

Code, copyright, or technical jargon

link to

No Comments

No Known Activations