INDEX

Explanations

acronyms related to the natural gas industry

oai_token-act-pair · gpt-3.5-turbo

references to natural gas

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Reggie

-0.73

 pepper

-0.72

 Danny

-0.69

 Betty

-0.65

 tabs

-0.65

Vac

-0.64

 Melvin

-0.64

 bart

-0.63

 Bart

-0.62

 causal

-0.62

POSITIVE LOGITS

NG

4.75

NG

1.50

ng

1.49

NN

1.40

NK

1.39

NR

1.32

NT

1.29

ND

1.28

PG

1.27

BG

1.21

Activations Density 0.013%

acronyms related to the natural gas industry

references to natural gas

No Comments

No Known Activations