INDEX

Explanations

mentions of toxic substances or situations

oai_token-act-pair · gpt-3.5-turbo

references to toxic substances and their effects

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

gain

-0.80

hung

-0.79

quart

-0.74

through

-0.73

FORE

-0.71

held

-0.71

Month

-0.71

telling

-0.70

stand

-0.70

UTERS

-0.70

POSITIVE LOGITS

 toxic

1.13

 masculinity

1.03

ologically

0.99

 toxicity

0.99

 substances

0.96

 poisoning

0.95

 fumes

0.95

ologist

0.94

ological

0.92

 poisonous

0.92

Activations Density 0.009%

mentions of toxic substances or situations

references to toxic substances and their effects

No Comments

No Known Activations