INDEX

Explanations

phrases related to legal and financial exemptions

oai_token-act-pair · gpt-3.5-turbo

phrases related to legal exemptions and liabilities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Characters

-0.74

Osw

-0.69

pire

-0.69

nice

-0.68

rame

-0.66

pires

-0.65

apsed

-0.65

Tool

-0.65

finished

-0.64

oled

-0.63

POSITIVE LOGITS

 liability

1.43

 taxation

1.42

 prosecution

1.40

 scrutiny

1.32

 penalties

1.31

 taxes

1.27

 antitrust

1.25

 prohibitions

1.18

 sanctions

1.16

 fines

1.14

Activations Density 0.229%

phrases related to legal and financial exemptions

phrases related to legal exemptions and liabilities

No Comments

No Known Activations