INDEX

Explanations

references to the concept of "more" or increases in quantity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

SNL

-0.73

(&:

-0.73

bufio

-0.71

 Caballero

-0.70

️

-0.70

❥

-0.69

 Cluj

-0.69

__()

-0.68

 africana

-0.66

Discografia

-0.66

POSITIVE LOGITS

 more

1.77

 MORE

1.60

more

1.53

 More

1.41

More

1.41

MORE

1.41

 Moreno

1.22

 Moreira

1.13

 Moreau

1.12

emore

1.12

Activations Density 0.147%

references to the concept of "more" or increases in quantity

No Comments

No Known Activations