INDEX

Explanations

references to quickness or efficiency

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 Faust

-0.70

šina

-0.69

 Gallardo

-0.69

ValueStyle

-0.69

 Senhora

-0.66

 חיצוניים

-0.66

ReusableCell

-0.65

 Nadel

-0.65

MathML

-0.64

 Sall

-0.64

POSITIVE LOGITS

 Brief

1.05

 QUICK

1.03

 Quick

1.02

 brief

0.99

Quick

0.95

QUICK

0.94

 quick

0.89

 Quicks

0.87

 Quis

0.86

quick

0.85

Activations Density 0.065%

references to quickness or efficiency

No Comments

No Known Activations