INDEX

Explanations

instances of the word "one" with varying degrees of certainty

oai_token-act-pair · gpt-3.5-turbo

references to obtaining or having a single item

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

grounds

-0.69

 Vaugh

-0.64

aeda

-0.64

ories

-0.59

osponsors

-0.58

inity

-0.58

 Ezek

-0.57

lov

-0.56

 margins

-0.56

veyard

-0.55

POSITIVE LOGITS

 sided

0.92

 hundred

0.90

 Hundred

0.86

 thousand

0.82

elf

0.82

of

0.82

 dimensional

0.79

Shot

0.76

 yourself

0.75

 thereof

0.74

Activations Density 0.115%

instances of the word "one" with varying degrees of certainty

references to obtaining or having a single item

No Comments

No Known Activations