INDEX

Explanations

phrases related to photographs or images being taken

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "taken" in the context of photographs or images

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

 loosen

-0.73

 pled

-0.73

 revocation

-0.72

 dues

-0.71

 punish

-0.70

regulated

-0.69

 regress

-0.68

 subsid

-0.67

estinal

-0.67

 clause

-0.66

POSITIVE LOGITS

 Thumbnails

0.94

Capture

0.89

 Mehran

0.89

 Aerial

0.87

 photographs

0.86

 Photographer

0.84

 photographers

0.84

phot

0.83

 photograp

0.82

 msec

0.81

Activations Density 0.102%

phrases related to photographs or images being taken

occurrences of the word "taken" in the context of photographs or images

No Comments

No Known Activations