INDEX

Explanations

phrases related to consequences, aftermath, or events that follow a particular action or incident

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "in the wake of."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

afort

-0.66

entric

-0.63

inates

-0.61

avorite

-0.61

 Spirits

-0.61

uca

-0.60

 recip

-0.60

rib

-0.60

venants

-0.59

 forgiven

-0.58

POSITIVE LOGITS

achine

0.98

fulness

0.86

mares

0.80

 wake

0.76

giving

0.75

DragonMagazine

0.74

manship

0.73

bringer

0.72

math

0.71

storm

0.71

Activations Density 0.011%

phrases related to consequences, aftermath, or events that follow a particular action or incident

instances of the phrase "in the wake of."

No Comments

No Known Activations