INDEX

Explanations

terms related to controversial events involving attacks, killings, or incidents

oai_token-act-pair · gpt-3.5-turbo

references to politically motivated incidents or conspiracy theories

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

Tokens

-0.91

anguages

-0.76

resh

-0.72

imei

-0.71

izons

-0.71

»Ĵ

-0.70

oise

-0.67

stores

-0.67

 worlds

-0.66

 Devi

-0.66

POSITIVE LOGITS

 deliberate

1.28

 spontaneous

1.25

 retaliation

1.20

 accidental

1.20

 intentional

1.17

 coincidence

1.14

 hoax

1.13

 provocation

1.12

 orchestrated

1.11

 unintentional

1.09

Activations Density 0.390%

terms related to controversial events involving attacks, killings, or incidents

references to politically motivated incidents or conspiracy theories

No Comments

No Known Activations