INDEX

Explanations

phrases that emphasize inclusion or membership in a group or category

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

 humaines

-0.66

 ſche

-0.66

($)

-0.64

lorette

-0.63

thalene

-0.62

 Philo

-0.60

 sandero

-0.60

 lepto

-0.58

Sola

-0.57

 Arno

-0.57

POSITIVE LOGITS

 including

1.42

 Including

1.36

including

1.29

Including

1.28

INCLUDING

1.26

 INCLUDING

1.24

 waaronder

1.16

 einschließlich

1.11

 включая

1.04

 incluindo

1.01

Activations Density 0.113%

phrases that emphasize inclusion or membership in a group or category

No Comments

No Known Activations