INDEX

Explanations

instances of the word "belief"

oai_token-act-pair · gpt-3.5-turbo

concepts related to belief systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

xon

-0.73

bye

-0.72

avis

-0.71

nice

-0.69

yna

-0.69

apeake

-0.64

eded

-0.64

mentioned

-0.63

edited

-0.63

 Sieg

-0.62

POSITIVE LOGITS

 belief

1.19

 beliefs

0.94

ieve

0.89

 conviction

0.87

 believing

0.83

 guiActiveUn

0.82

 adherence

0.81

 Belief

0.81

 disbelief

0.80

fulness

0.79

Activations Density 0.013%

instances of the word "belief"

concepts related to belief systems

No Comments

No Known Activations