INDEX

Explanations

terms related to control, targeting, and specific actions

oai_token-act-pair · gpt-3.5-turbo

terms related to intentional and systematic actions or experiments

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

pires

-0.84

aughty

-0.79

orthy

-0.77

asio

-0.74

baum

-0.71

 Qiao

-0.70

kj

-0.70

astical

-0.69

 Carbuncle

-0.69

 esteemed

-0.69

POSITIVE LOGITS

 demolition

1.24

 elimination

1.24

 executions

1.23

 deletion

1.21

 removal

1.15

 assass

1.14

 eviction

1.13

 destruction

1.13

 demol

1.13

 killings

1.13

Activations Density 0.226%

terms related to control, targeting, and specific actions

terms related to intentional and systematic actions or experiments

No Comments

No Known Activations