INDEX

Explanations

phrases indicating a suggestion or command for action

oai_token-act-pair · gpt-3.5-turbo

the phrase "better" in various contexts indicating advice or recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

eur

-0.73

Nob

-0.72

esville

-0.72

etic

-0.70

untarily

-0.67

urous

-0.66

oidal

-0.62

éĥ

-0.61

MG

-0.61

itory

-0.61

POSITIVE LOGITS

 behaved

0.80

 suited

0.78

 luck

0.76

ment

0.75

 manners

0.72

 beware

0.72

 than

0.71

idge

0.70

ments

0.69

 lett

0.66

Activations Density 0.040%

phrases indicating a suggestion or command for action

the phrase "better" in various contexts indicating advice or recommendations

No Comments

No Known Activations