INDEX

Explanations

instances where the word "to" is followed by another word that completes a particular action or transaction

oai_token-act-pair · gpt-3.5-turbo

phrases indicating monetary transactions or payments

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

limits

-0.67

typ

-0.66

aign

-0.63

furt

-0.63

ational

-0.63

spection

-0.61

ItemThumbnailImage

-0.61

growing

-0.61

felt

-0.60

umerable

-0.60

POSITIVE LOGITS

 compensate

1.08

 acquire

1.05

 settle

1.02

 advertise

1.02

 maintain

1.00

 defend

0.93

 upkeep

0.92

 operate

0.91

 participate

0.90

 incarcer

0.89

Activations Density 0.116%

instances where the word "to" is followed by another word that completes a particular action or transaction

phrases indicating monetary transactions or payments

No Comments

No Known Activations