INDEX

Explanations

mentions of political figures and activities regarding potential collusion between political campaigns and foreign governments, specifically Russia

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

Data Type

torch.float32

Hook Point

blocks.8.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

 ç¥ŀ

-0.95

enment

-0.84

}}}

-0.81

iku

-0.81

ciation

-0.81

vation

-0.81

 univers

-0.79

 roar

-0.77

liga

-0.76

ãĤ¦ãĤ¹

-0.75

POSITIVE LOGITS

 clandestine

1.24

 covert

1.23

 allegedly

1.23

 illicit

1.21

 alleged

1.19

CIA

1.18

 espionage

1.15

 Manafort

1.15

 secretly

1.13

 bribes

1.13

Activations Density 3.866%

mentions of political figures and activities regarding potential collusion between political campaigns and foreign governments, specifically Russia

No Comments

No Known Activations