INDEX

Explanations

phrases related to US Presidents Trump and Obama

oai_token-act-pair · gpt-3.5-turbo

linguistic patterns or structures that denote possession or association in relation to individuals, particularly in political contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 subdivision

-0.71

 neutron

-0.67

 filler

-0.66

 Proto

-0.66

 Bulgarian

-0.66

 Moor

-0.65

 Norn

-0.63

 Finnish

-0.63

Owl

-0.62

aimon

-0.62

POSITIVE LOGITS

realDonaldTrump

1.33

Jr

1.10

1.03

Obama

0.98

administ

0.95

Clinton

0.93

TRUMP

0.92

ï

0.91

Romney

0.90

Trump

0.90

Activations Density 0.163%

phrases related to US Presidents Trump and Obama

linguistic patterns or structures that denote possession or association in relation to individuals, particularly in political contexts

No Comments

No Known Activations