INDEX

Explanations

phrases or sentences that indicate a change or transition is happening

oai_token-act-pair · gpt-3.5-turbo

the word "soon" and its variations, indicating imminent action or events

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

ocker

-0.68

utenberg

-0.67

atively

-0.66

BIP

-0.63

urized

-0.61

che

-0.60

retched

-0.59

rament

-0.59

agate

-0.59

ribe

-0.58

POSITIVE LOGITS

 thereafter

1.03

aneously

0.88

 enough

0.82

 arrives

0.75

 soon

0.74

 Soon

0.74

Soon

0.73

 afterwards

0.72

 approaching

0.71

 surpass

0.70

Activations Density 0.017%

phrases or sentences that indicate a change or transition is happening

the word "soon" and its variations, indicating imminent action or events

No Comments

No Known Activations