INDEX

Explanations

legal terms and mentions of government entities in the United States

oai_token-act-pair · gpt-3.5-turbo

references to the United States government and its various agencies or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

lets

-0.84

bats

-0.82

zbollah

-0.79

ebook

-0.74

igs

-0.69

bably

-0.69

ãĤ·

-0.69

zzle

-0.68

bes

-0.68

issance

-0.67

POSITIVE LOGITS

 Department

1.01

 Marshal

1.00

 Attorney

0.94

 Postal

0.94

 Citizenship

0.94

 Justice

0.94

GI

0.93

 Cellular

0.91

 Marsh

0.90

 Territories

0.88

Activations Density 0.080%

legal terms and mentions of government entities in the United States

references to the United States government and its various agencies or actions

No Comments

No Known Activations