INDEX

Explanations

phrases related to political figures, specifically mentions of President Donald Trump

oai_token-act-pair · gpt-3.5-turbo

frequent mentions of Donald Trump

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

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

 Mortal

-0.75

gnu

-0.68

Offline

-0.67

çĶ°

-0.67

 Emacs

-0.64

IE

-0.61

BILITY

-0.61

ASA

-0.61

 Shutterstock

-0.61

ANA

-0.61

POSITIVE LOGITS

care

0.98

 Donald

0.93

 appoint

0.88

Jr

0.84

 Tower

0.83

 tweeted

0.83

Taj

0.83

Care

0.81

omics

0.80

Trump

0.79

Activations Density 0.057%

phrases related to political figures, specifically mentions of President Donald Trump

frequent mentions of Donald Trump

No Comments

No Known Activations