INDEX

Explanations

references to specific legal sections and chemical compounds

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

structured alphanumeric tokens typical of formal technical text, such as legal citations, scientific/chemical nomenclature, and dotted version numbers.

oai_token-act-pair · gpt-5 Triggered by @arielgoldsteinlab

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 lipat

-0.37

 amigas

-0.35

 rueda

-0.33

 rallado

-0.33

 Italij

-0.33

 Secretario

-0.33

 testigo

-0.33

 criatura

-0.32

 accesorios

-0.30

 organes

-0.29

POSITIVE LOGITS

ModelAdmin

0.76

 EconPapers

0.71

 pinulongan

0.69

intios

0.67

mybatisplus

0.67

makeConstraints

0.66

Rohy

0.66

elemField

0.65

 AssemblyCulture

0.63

RegressionTest

0.63

Activations Density 1.722%

references to specific legal sections and chemical compounds

structured alphanumeric tokens typical of formal technical text, such as legal citations, scientific/chemical nomenclature, and dotted version numbers.

No Comments

No Known Activations