INDEX

Explanations

references to "post" followed by numbers indicating significant events or eras

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

words related to events or conditions occurring after a specific time or event

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Qiao

-0.84

 Orchestra

-0.74

 Bett

-0.73

ï¸

-0.73

 bands

-0.72

owship

-0.70

Ø©

-0.69

 Beans

-0.65

 Bars

-0.65

 Skin

-0.65

POSITIVE LOGITS

mortem

1.47

doctoral

1.31

gres

1.31

modern

1.27

hum

1.26

natal

1.22

graduate

1.20

cards

1.15

traumatic

1.13

colonial

1.11

Activations Density 0.036%

references to "post" followed by numbers indicating significant events or eras

words related to events or conditions occurring after a specific time or event

No Comments

No Known Activations