INDEX

Explanations

references to organizations and their affiliations within various contexts

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

which leads to

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

WebElementEntity

-0.81

PostInfinity

-0.73

новниш

-0.70

 esternos

-0.67

 مرئيه

-0.67

IUrlHelper

-0.67

 initComponents

-0.66

RegressionTest

-0.65

ніципалі

-0.65

XmlAccessorType

-0.64

POSITIVE LOGITS

 which

0.49

 itself

0.35

 being

0.34

 welke

0.33

 vilken

0.32

 welches

0.31

강

0.31

 serial

0.31

 were

0.30

 revolution

0.29

Activations Density 0.042%

references to organizations and their affiliations within various contexts

which leads to

No Comments

No Known Activations