INDEX

Explanations

phrases related to readings or interpretations of data

oai_token-act-pair · gpt-3.5-turbo

terms and phrases related to readings and observations

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

 grooming

-0.70

 burying

-0.69

 bully

-0.69

ijk

-0.67

 vanity

-0.67

 Gron

-0.66

 McGr

-0.66

 stamp

-0.65

 butt

-0.65

 camel

-0.64

POSITIVE LOGITS

 readings

2.54

heard

2.07

seen

1.69

song

1.56

saw

1.29

 Gathering

1.06

orders

0.99

warning

0.96

adiq

0.95

 screenings

0.94

Activations Density 0.045%

phrases related to readings or interpretations of data

terms and phrases related to readings and observations

No Comments

No Known Activations