INDEX

Explanations

religious terms or references

oai_token-act-pair · gpt-3.5-turbo

terms related to religion and spirituality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 wide

-0.69

HAEL

-0.67

 Mercury

-0.65

 Wilde

-0.63

 Marketplace

-0.63

 wider

-0.60

 furthe

-0.60

 thinly

-0.59

 Walters

-0.58

 slate

-0.58

POSITIVE LOGITS

igion

1.61

iever

1.48

iable

1.45

iability

1.41

iance

1.39

ief

1.38

atively

1.32

ievers

1.32

ieved

1.29

ieving

1.29

Activations Density 0.022%

religious terms or references

terms related to religion and spirituality

No Comments

No Known Activations