INDEX

Explanations

emails and their related terms

oai_token-act-pair · gpt-3.5-turbo

references to emails

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Osaka

-0.72

iculture

-0.69

LOS

-0.68

axis

-0.67

ESCO

-0.66

INAL

-0.66

oba

-0.65

@#&

-0.65

 Flags

-0.63

vic

-0.63

POSITIVE LOGITS

 exchanged

0.99

 dumps

0.98

 Emails

0.97

 dump

0.94

 Podesta

0.94

 correspondence

0.92

 leaked

0.91

ileaks

0.89

 emails

0.86

archive

0.86

Activations Density 0.042%

emails and their related terms

references to emails

No Comments

No Known Activations