INDEX

Explanations

photographs or terms related to them

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Photograph" or related variations indicating images or photographic content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 anyways

-0.72

 goodness

-0.70

oku

-0.67

 evil

-0.67

 advancement

-0.66

 happ

-0.62

 realm

-0.59

 sanity

-0.59

 future

-0.59

 spew

-0.58

POSITIVE LOGITS

 Photograph

4.44

Phot

1.82

 photograph

1.73

 Photographer

1.63

 photographs

1.58

 photographed

1.54

 PHOTO

1.52

 Photos

1.48

 Photo

1.48

 Photography

1.39

Activations Density 0.008%

photographs or terms related to them

occurrences of the word "Photograph" or related variations indicating images or photographic content

No Comments

No Known Activations