INDEX

Explanations

images or photos mentioned in text

oai_token-act-pair · gpt-3.5-turbo

occurrences of the phrase "photo of" followed by various subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

hement

-0.81

wcs

-0.80

 rely

-0.71

aucuses

-0.71

 ende

-0.70

 relied

-0.68

 levied

-0.68

 depended

-0.67

ktop

-0.66

 pact

-0.64

POSITIVE LOGITS

 images

0.66

 Images

0.66

 Photography

0.66

 mine

0.66

 them

0.64

milo

0.63

 footage

0.63

 course

0.63

him

0.63

gery

0.62

Activations Density 0.163%

images or photos mentioned in text

occurrences of the phrase "photo of" followed by various subjects

No Comments

No Known Activations