INDEX

Explanations

references to documents, articles, or statements, indicating their origin or context

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

research papers or studies

np_acts-logits-general · gemini-2.0-flash

this publication or work

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 незавершена

-0.69

:][

-0.67

Personendaten

-0.65

AnchorTagHelper

-0.65

 nakalista

-0.60

 suivante

-0.60

 defaultstate

-0.59

 referenties

-0.59

Odkazy

-0.58

umumkan

-0.58

POSITIVE LOGITS

 paper

0.60

 page

0.58

 document

0.54

DataAnnotations

0.53

 report

0.52

ruptedException

0.47

DOCTYPE

0.47

 chapter

0.45

 program

0.45

 letter

0.45

Activations Density 0.283%

references to documents, articles, or statements, indicating their origin or context

research papers or studies

this publication or work

No Comments

No Known Activations