INDEX

Explanations

proper nouns associated with a specific format, possibly dates of birth or events in parentheses

oai_token-act-pair · gpt-3.5-turbo

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

 bump

-0.77

 equival

-0.76

 increment

-0.75

 transact

-0.75

 treat

-0.75

 exporting

-0.74

 everyday

-0.74

 trivial

-0.74

 mutually

-0.73

 lull

-0.73

POSITIVE LOGITS

pictured

1.72

pron

1.44

aka

1.42

whose

1.34

formerly

1.34

who

1.30

sic

1.28

left

1.27

above

1.24

shown

1.22

Activations Density 0.099%

proper nouns associated with a specific format, possibly dates of birth or events in parentheses

No Comments

No Known Activations