INDEX

Explanations

words related to privacy and potentially sensitive information

oai_token-act-pair · gpt-3.5-turbo

adverbs that imply a state of legitimacy, regularity, or frequency

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Payne

-0.70

arella

-0.67

anmar

-0.66

Voy

-0.65

quickShipAvailable

-0.65

oire

-0.65

 Timbers

-0.64

 Laos

-0.63

ogue

-0.63

raltar

-0.62

POSITIVE LOGITS

 spaced

0.85

 differentiated

0.81

accessible

0.81

protected

0.80

 identifiable

0.80

 populated

0.80

putable

0.79

 formulated

0.79

 defined

0.77

 constructed

0.77

Activations Density 0.216%

words related to privacy and potentially sensitive information

adverbs that imply a state of legitimacy, regularity, or frequency

No Comments

No Known Activations