INDEX

Explanations

prefixes followed by an adjective

oai_token-act-pair · gpt-3.5-turbo

variations of the prefix "un-" indicating negation or oppositeness

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

 Barker

-0.76

 circle

-0.66

 Christy

-0.63

 Ibrahim

-0.63

minster

-0.62

SHIP

-0.60

 Khan

-0.59

 Lambert

-0.57

 Warlock

-0.56

 deprivation

-0.56

POSITIVE LOGITS

akable

1.16

cedented

1.03

oppable

1.02

ailability

0.97

ached

0.91

ailable

0.87

orable

0.87

ishable

0.86

ifiable

0.83

athed

0.82

Activations Density 0.061%

prefixes followed by an adjective

variations of the prefix "un-" indicating negation or oppositeness

No Comments

No Known Activations