INDEX

Explanations

instances where an action is prohibited or restricted

oai_token-act-pair · gpt-3.5-turbo

terms related to restrictions or bans

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

eon

-0.74

arger

-0.74

temp

-0.72

along

-0.69

Roy

-0.67

illation

-0.67

LV

-0.66

ensional

-0.65

Transform

-0.65

PE

-0.64

POSITIVE LOGITS

 prohibited

1.28

etheless

1.05

 forbidden

0.98

bidden

0.90

 forbids

0.85

 prohibits

0.85

wana

0.81

avorite

0.80

 permitted

0.79

 tradem

0.78

Activations Density 0.012%

instances where an action is prohibited or restricted

terms related to restrictions or bans

No Comments

No Known Activations