INDEX

Explanations

references to historical events or incidents involving criminal activities

oai_token-act-pair · gpt-3.5-turbo

references to significant historical events, particularly related to terrorism and major attacks

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

heit

-0.94

SPONSORED

-0.92

AppData

-0.88

icip

-0.83

hm

-0.82

ometers

-0.82

imentary

-0.81

sonian

-0.79

ictionary

-0.79

reau

-0.75

POSITIVE LOGITS

 massacre

1.32

 disappearance

1.20

 assassination

1.20

 killings

1.15

 deadliest

1.15

 deaths

1.13

 massacres

1.12

 shootings

1.11

 Holocaust

1.09

 atroc

1.09

Activations Density 0.310%

references to historical events or incidents involving criminal activities

references to significant historical events, particularly related to terrorism and major attacks

No Comments

No Known Activations