INDEX

Explanations

references to specific studies and instances in a research or clinical context

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

single instance or count

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

mphs

-0.40

 Económica

-0.34

 sagrado

-0.34

躇

-0.33

 Divers

-0.33

makeConstraints

-0.32

ższe

-0.32

 Públicas

-0.31

 ragazzi

-0.31

 affari

-0.31

POSITIVE LOGITS

 defaultstate

0.70

bootstrapcdn

0.66

 للمعارف

0.64

CppMethod

0.63

 autorytatywna

0.62

 незавершена

0.61

RegressionTest

0.59

":[{

0.58

 PeEnEo

0.57

 ویکی‌پدی

0.56

Activations Density 0.060%

references to specific studies and instances in a research or clinical context

single instance or count

No Comments

No Known Activations