INDEX

Explanations

phrases related to official ceremonies or legal procedures

oai_token-act-pair · gpt-3.5-turbo

references to the act of being sworn in or taking an oath

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 guiActiveUnfocused

-0.69

nesota

-0.69

Mind

-0.69

edia

-0.67

negie

-0.65

Gray

-0.65

Stage

-0.64

 Giov

-0.64

anon

-0.62

plex

-0.62

POSITIVE LOGITS

 sworn

1.27

 oath

1.14

 swearing

0.99

 swore

0.86

 swear

0.81

iffs

0.79

 duty

0.76

 vows

0.75

 interpreter

0.74

aith

0.73

Activations Density 0.006%

phrases related to official ceremonies or legal procedures

references to the act of being sworn in or taking an oath

No Comments

No Known Activations