INDEX

Explanations

phrases related to political extremism and conflicts

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 thru

-1.03

[/

-0.87

 blah

-0.87

???

-0.85

uh

-0.84

!!!!

-0.83

???

-0.81

 crappy

-0.81

 ONLY

-0.81

 shitty

-0.80

POSITIVE LOGITS

 Nato

1.10

 Britons

1.10

 Rouhani

1.07

 Ukip

1.06

 organised

1.04

 Kazakh

1.04

 organis

1.03

 Islamists

1.00

 migrants

1.00

 diplomats

0.99

Activations Density 1.680%

phrases related to political extremism and conflicts

No Comments

No Known Activations