INDEX

Explanations

references to government agencies related to law enforcement and immigration control

oai_token-act-pair · gpt-3.5-turbo

terms related to immigration and customs enforcement agencies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

icular

-0.87

asted

-0.84

bed

-0.80

heading

-0.76

alysed

-0.76

ighty

-0.74

outine

-0.74

rower

-0.72

ebin

-0.72

ocl

-0.71

POSITIVE LOGITS

 Blocks

1.07

 Accountability

1.07

 Bureau

1.03

 Affairs

1.03

 Services

1.01

 Resources

1.00

 Association

1.00

 Institution

0.99

 Rights

0.98

 Cooperative

0.98

Activations Density 0.063%

references to government agencies related to law enforcement and immigration control

terms related to immigration and customs enforcement agencies

No Comments

No Known Activations