INDEX

Explanations

references to directional positions or sections within a layout

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

addCriterion

-0.65

 Infórmanos

-0.59

verifyException

-0.57

 AssemblyCompany

-0.55

oporosis

-0.53

modelBuilder

-0.53

AnimationsModule

-0.52

PoolSize

-0.52

pulseira

-0.52

MeasureSpec

-0.52

POSITIVE LOGITS

Left

1.22

 Left

1.18

left

1.18

 left

1.18

LEFT

1.02

 LEFT

1.02

setLeft

0.96

 izquierdo

0.86

Lef

0.85

pleft

0.84

Activations Density 0.077%

references to directional positions or sections within a layout

No Comments

No Known Activations