INDEX

Explanations

technical terms and concepts related to data analysis and processing

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

smoothing, scaling, and error

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

rungsseite

-0.69

adaptiveStyles

-0.53

 oprot

-0.52

 nakalista

-0.52

Географиясе

-0.51

AccessFile

-0.49

ngdoc

-0.49

UrlResolution

-0.47

ialog

-0.47

 Normdatei

-0.47

POSITIVE LOGITS

成

0.41

big

0.40

Big

0.39

 aggregate

0.39

 coarser

0.38

GEBURTSDATUM

0.37

 coarse

0.37

Organisateur

0.37

Aggregate

0.36

IZONT

0.36

Activations Density 0.106%

technical terms and concepts related to data analysis and processing

smoothing, scaling, and error

No Comments

No Known Activations