INDEX

Explanations

emphatic expressions or words conveying strong emotions

oai_token-act-pair · gpt-3.5-turbo

exclamatory or emphatic expressions of frustration or intensity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Flavoring

-0.70

Comment

-0.60

////

-0.60

ynt

-0.59

":[{"

-0.59

yan

-0.59

Remove

-0.58

occup

-0.57

 additionally

-0.57

Serv

-0.56

POSITIVE LOGITS

near

0.94

ibly

0.80

holes

0.78

ation

0.77

 kidding

0.76

aughtered

0.74

loads

0.70

 liar

0.69

rolled

0.69

 hurry

0.68

Activations Density 0.069%

emphatic expressions or words conveying strong emotions

exclamatory or emphatic expressions of frustration or intensity

No Comments

No Known Activations