INDEX

Explanations

prepositions that indicate the use of something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "to" in various contexts

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

 Needs

-0.73

icipated

-0.66

 Niet

-0.66

ENTION

-0.65

tions

-0.64

 Creed

-0.63

listed

-0.63

 Removal

-0.61

 Horton

-0.60

soDeliveryDate

-0.60

POSITIVE LOGITS

 promote

1.32

 enhance

1.31

 propel

1.30

 denote

1.26

 justify

1.25

 motivate

1.25

 generate

1.25

 bolster

1.24

 communicate

1.24

 create

1.21

Activations Density 0.137%

prepositions that indicate the use of something

instances of the word "to" in various contexts

No Comments

No Known Activations