INDEX

Explanations

phrases related to doing something in consideration or benefit of a particular reason

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize the reasons or justifications for actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

iste

-0.83

igue

-0.72

quartered

-0.69

glass

-0.69

 Cosponsors

-0.68

Jar

-0.67

falls

-0.67

til

-0.65

ned

-0.63

Lev

-0.61

POSITIVE LOGITS

 preserving

1.03

 sanity

0.93

 protecting

0.88

 keeping

0.85

 humanity

0.82

 avoiding

0.81

 sake

0.81

 preservation

0.77

ĺħ

0.77

 Humanity

0.77

Activations Density 0.091%

phrases related to doing something in consideration or benefit of a particular reason

phrases that emphasize the reasons or justifications for actions

No Comments

No Known Activations