INDEX

Explanations

references to or mentions of Donald Trump

oai_token-act-pair · gpt-3.5-turbo

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 @ 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

à¨

-0.70

owship

-0.69

Sov

-0.68

MpServer

-0.63

 Isles

-0.61

CAP

-0.61

INAL

-0.61

 intrusion

-0.60

Bird

-0.60

Products

-0.59

POSITIVE LOGITS

Jr

1.15

 Donald

1.15

(@

0.96

 campaigned

0.96

Jr

0.95

 speaks

0.95

 greets

0.89

 retweet

0.85

 listens

0.83

Sr

0.82

Activations Density 0.097%

references to or mentions of Donald Trump

mentions of Donald Trump

No Comments

No Known Activations