INDEX

Explanations

phrases related to wealth and power

oai_token-act-pair · gpt-3.5-turbo

themes related to wealth and suffering

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 ACTIONS

-0.81

EVA

-0.80

 Autism

-0.76

 trending

-0.76

 Nieto

-0.74

 Surveillance

-0.71

 Spotlight

-0.71

INTON

-0.69

 Canaver

-0.69

 Agenda

-0.69

POSITIVE LOGITS

 labou

0.88

fare

0.85

folk

0.84

cloth

0.83

,—

0.82

 Zoro

0.81

 consec

0.81

ox

0.78

 precept

0.77

 armour

0.77

Activations Density 0.706%

phrases related to wealth and power

themes related to wealth and suffering

No Comments

No Known Activations