INDEX

Explanations

phrases related to medical and scientific studies or observations

oai_token-act-pair · gpt-3.5-turbo

opening parentheses in the text

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

quished

-0.73

 Lumpur

-0.70

icy

-0.66

 tear

-0.62

 rede

-0.61

 reneg

-0.61

rem

-0.61

wre

-0.60

 resh

-0.60

 lull

-0.60

POSITIVE LOGITS

see

1.49

sic

1.34

including

1.31

Figure

1.27

emphasis

1.20

approximately

1.19

excluding

1.18

typically

1.18

especially

1.17

both

1.17

Activations Density 0.175%

phrases related to medical and scientific studies or observations

opening parentheses in the text

No Comments

No Known Activations