INDEX

Explanations

religious terms or references

oai_token-act-pair · gpt-3.5-turbo

terms associated with belief and believers in a religious context

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anka

-0.67

 parks

-0.66

 Mamm

-0.64

 Armour

-0.63

æ©Ł

-0.62

 Paste

-0.62

 Mane

-0.62

 Grab

-0.61

 Plaza

-0.59

 Agric

-0.59

POSITIVE LOGITS

ievers

1.26

ieving

1.04

 believer

1.01

ieve

1.00

iever

0.94

cius

0.91

 believers

0.87

ieved

0.83

ieves

0.83

 believing

0.79

Activations Density 0.025%

religious terms or references

terms associated with belief and believers in a religious context

No Comments

No Known Activations