INDEX

Explanations

phrases that discuss the concept of being treated or handled in various ways or under specific conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 Româ

-0.55

 bénéficiaire

-0.55

 Constitucional

-0.54

 potest

-0.54

belec

-0.53

 assise

-0.52

 originaire

-0.50

 Económica

-0.50

 Inscription

-0.50

 alas

-0.50

POSITIVE LOGITS

 treated

1.08

 treating

1.03

treated

1.03

 treat

0.99

 treatment

0.94

 behave

0.93

Treating

0.91

treatment

0.91

 Treat

0.90

 behaved

0.89

Activations Density 0.345%

phrases that discuss the concept of being treated or handled in various ways or under specific conditions

No Comments

No Known Activations