INDEX

Explanations

phrases related to sharing one's opinion or experience publicly

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate speaking out or expressing opinions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Carbuncle

-0.88

plet

-0.74

ipment

-0.70

 Landing

-0.65

CG

-0.64

 refresh

-0.63

 Reincarn

-0.61

 Purg

-0.61

 Associates

-0.61

atomic

-0.61

POSITIVE LOGITS

 against

1.49

against

1.29

 forcefully

1.23

 Against

1.20

 louder

1.20

 loudly

1.15

 publicly

1.11

Against

1.06

 anonymously

1.06

voc

1.05

Activations Density 0.132%

phrases related to sharing one's opinion or experience publicly

phrases that indicate speaking out or expressing opinions

No Comments

No Known Activations