INDEX

Explanations

phrases related to technology and user experience

oai_token-act-pair · gpt-3.5-turbo

expressions related to 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Osama

-0.62

ÙĬ

-0.60

 ÂŃ

-0.59

aughtered

-0.58

 Rodham

-0.58

='

-0.57

Ùħ

-0.56

Ø§ÙĦ

-0.55

 Gulf

-0.55

terrorist

-0.54

POSITIVE LOGITS

 downside

0.82

 drawback

0.64

 strengths

0.60

 lazy

0.58

 disclaimer

0.57

 honestly

0.57

laz

0.56

 disadvantages

0.56

 setups

0.54

 devs

0.54

Activations Density 1.993%

phrases related to technology and user experience

expressions related to advice or recommendations

No Comments

No Known Activations