INDEX

Explanations

instances of the word "up" or related directional terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 zoude

-0.80

 plufieurs

-0.77

 ſind

-0.75

 surla

-0.72

 HasFactory

-0.71

 geſ

-0.70

 müſſen

-0.68

HideFlags

-0.68

 laſſen

-0.68

 Numerade

-0.68

POSITIVE LOGITS

up

2.08

 down

1.30

off

1.19

 away

1.02

out

0.90

 back

0.85

 jusqu

0.77

 into

0.74

 along

0.70

 upto

0.65

Activations Density 0.211%

instances of the word "up" or related directional terms

No Comments

No Known Activations