INDEX

Explanations

phrases related to health and well-being

oai_token-act-pair · gpt-3.5-turbo

concepts related to health, well-being, and safety

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

isSpecialOrderable

-0.85

},{"

-0.62

zan

-0.61

 =================================

-0.61

"},{"

-0.57

roo

-0.57

 Secondly

-0.57

iolet

-0.56

 Shades

-0.54

 underestimated

-0.53

POSITIVE LOGITS

 workings

0.93

liest

0.92

 fortunes

0.88

 whereabouts

0.80

iest

0.80

 birthplace

0.79

of

0.78

inations

0.76

 lives

0.75

 merits

0.75

Activations Density 0.692%

phrases related to health and well-being

concepts related to health, well-being, and safety

No Comments

No Known Activations