INDEX

Explanations

phrases indicating a contradiction or opposing viewpoint

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize a contrast or contradiction

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

soDeliveryDate

-0.76

esh

-0.75

iguous

-0.71

ected

-0.70

rosso

-0.68

walking

-0.64

 Takes

-0.63

nin

-0.62

ãģ¦

-0.62

together

-0.62

POSITIVE LOGITS

 extent

1.74

 detriment

1.47

 tune

1.38

 contrary

1.26

 fullest

1.26

 exclusion

1.20

 degree

1.06

 point

1.05

venge

1.03

rouse

0.99

Activations Density 0.080%

phrases indicating a contradiction or opposing viewpoint

phrases that emphasize a contrast or contradiction

No Comments

No Known Activations