INDEX

Explanations

phrases related to decision-making or actions

oai_token-act-pair · gpt-3.5-turbo

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

emale

-0.70

 Chau

-0.66

idal

-0.63

acial

-0.63

Mau

-0.60

runner

-0.60

 deployed

-0.59

alde

-0.59

eded

-0.58

ãĤ¼ãĤ¦ãĤ¹

-0.58

POSITIVE LOGITS

something

1.58

 things

1.49

nothing

1.47

things

1.46

anything

1.46

Nothing

1.44

 Things

1.42

 Something

1.42

 something

1.40

 Anything

1.39

Activations Density 0.317%

phrases related to decision-making or actions

No Comments

No Known Activations