INDEX

Explanations

references to vertical and horizontal dimensions and orientations

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

gatsby

-0.76

:✨

-0.71

afficheront

-0.69

BeginContext

-0.68

 ***!

-0.67

strains

-0.67

atsby

-0.65

Idd

-0.65

TypedDataSet

-0.65

 kiện

-0.64

POSITIVE LOGITS

 horizontally

1.27

 horizontal

1.27

Horizontal

1.25

 Horizontal

1.22

 vertically

1.21

Vertical

1.15

ticals

1.13

 vertical

1.12

 Vertical

1.11

horizontal

1.10

Activations Density 0.168%

references to vertical and horizontal dimensions and orientations

No Comments

No Known Activations