INDEX

Explanations

technology-related terms, focusing on voice-activated assistants such as Siri and Alexa

oai_token-act-pair · gpt-3.5-turbo

references to voice-activated assistants and their functionalities

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

iatus

-0.80

stead

-0.77

emate

-0.75

POS

-0.74

papers

-0.73

 Sabbath

-0.72

itted

-0.71

iker

-0.71

eways

-0.70

rug

-0.69

POSITIVE LOGITS

 Siri

1.04

 Cortana

1.00

Voice

0.81

earable

0.81

 Bots

0.76

 Alexa

0.74

 Enabled

0.71

 speaker

0.69

vana

0.69

 voice

0.69

Activations Density 0.027%

technology-related terms, focusing on voice-activated assistants such as Siri and Alexa

references to voice-activated assistants and their functionalities

No Comments

No Known Activations