INDEX

Explanations

words related to conflict and war, particularly in the context of East Ukraine and Russia accusations

oai_token-act-pair · gpt-3.5-turbo

phrases related to conflict and violence in Ukraine

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Barker

-0.96

 Mecca

-0.95

 Hayward

-0.93

GHC

-0.93

 Jinn

-0.92

 Dawkins

-0.91

Sox

-0.91

 java

-0.89

 scient

-0.85

Qur

-0.83

POSITIVE LOGITS

 Kiev

2.25

 Ukrainian

2.23

 Ukrainians

2.20

 Donetsk

2.15

Ukraine

2.10

 Ukraine

2.05

 Ukrain

1.93

 Yanukovych

1.85

Uk

1.82

 Crimea

1.81

Activations Density 0.426%

words related to conflict and war, particularly in the context of East Ukraine and Russia accusations

phrases related to conflict and violence in Ukraine

No Comments

No Known Activations